yaojingang
845 lines
95 KiB
HTML
845 lines
95 KiB
HTML
<!DOCTYPE html>
|
|
<html lang="zh-CN">
|
|
<head>
|
|
<meta charset="utf-8">
|
|
<meta name="viewport" content="width=device-width, initial-scale=1">
|
|
<title>Yao Meta Skill Review Studio 2.0</title>
|
|
<style>
|
|
:root {
|
|
--ink: #1B365D;
|
|
--text: #24201d;
|
|
--muted: #746d66;
|
|
--line: #e7ded2;
|
|
--soft: #faf8f5;
|
|
--pass: #1e6b52;
|
|
--warn: #9a6718;
|
|
--block: #9b2c2c;
|
|
}
|
|
* { box-sizing: border-box; }
|
|
html { scroll-behavior: smooth; }
|
|
body {
|
|
margin: 0;
|
|
background: #ffffff;
|
|
color: var(--text);
|
|
font-family: Georgia, "Times New Roman", "Songti SC", serif;
|
|
line-height: 1.58;
|
|
}
|
|
nav {
|
|
position: sticky;
|
|
top: 0;
|
|
z-index: 10;
|
|
display: flex;
|
|
gap: 4px;
|
|
justify-content: center;
|
|
flex-wrap: wrap;
|
|
padding: 10px 16px;
|
|
background: rgba(255,255,255,0.94);
|
|
border-bottom: 1px solid var(--line);
|
|
backdrop-filter: blur(10px);
|
|
}
|
|
nav a {
|
|
color: var(--ink);
|
|
text-decoration: none;
|
|
font-size: 14px;
|
|
padding: 7px 10px;
|
|
border-radius: 6px;
|
|
}
|
|
nav a:hover { background: var(--soft); }
|
|
main { max-width: 1180px; margin: 0 auto; padding: 44px 28px 76px; }
|
|
header { border-bottom: 1px solid var(--line); padding-bottom: 28px; margin-bottom: 28px; }
|
|
.eyebrow { color: var(--ink); font-size: 14px; letter-spacing: .08em; text-transform: uppercase; }
|
|
h1, h2, h3 { color: var(--text); font-weight: 500; margin: 0; letter-spacing: 0; }
|
|
h1 { font-size: clamp(34px, 5vw, 64px); line-height: 1.03; max-width: 920px; margin-top: 12px; }
|
|
h2 { font-size: 30px; margin-bottom: 14px; }
|
|
h3 { font-size: 19px; }
|
|
p { margin: 0; }
|
|
.lede { max-width: 820px; color: var(--muted); font-size: 20px; margin-top: 18px; }
|
|
.decision {
|
|
display: inline-flex;
|
|
align-items: baseline;
|
|
gap: 12px;
|
|
margin-top: 24px;
|
|
padding: 12px 16px;
|
|
border: 1px solid var(--line);
|
|
border-radius: 8px;
|
|
color: var(--ink);
|
|
background: var(--soft);
|
|
}
|
|
.decision strong { font-size: 28px; }
|
|
section { padding: 30px 0; border-bottom: 1px solid var(--line); scroll-margin-top: 76px; }
|
|
.metrics, .gates {
|
|
display: grid;
|
|
grid-template-columns: repeat(5, minmax(0, 1fr));
|
|
gap: 14px;
|
|
}
|
|
.gates { grid-template-columns: repeat(4, minmax(0, 1fr)); }
|
|
.metric, .gate {
|
|
border: 1px solid var(--line);
|
|
border-radius: 8px;
|
|
padding: 16px;
|
|
background: #fff;
|
|
min-width: 0;
|
|
}
|
|
.metric span, .gate span { display: block; color: var(--muted); font-size: 13px; }
|
|
.metric strong { display: block; color: var(--ink); font-size: 34px; line-height: 1.1; margin: 8px 0; }
|
|
.metric p, .gate p, .gate footer, .issues span, .evidence span { color: var(--muted); font-size: 14px; overflow-wrap: anywhere; }
|
|
.gate { display: flex; flex-direction: column; gap: 10px; }
|
|
.gate.pass { border-top: 4px solid var(--pass); }
|
|
.gate.warn { border-top: 4px solid var(--warn); }
|
|
.gate.block { border-top: 4px solid var(--block); }
|
|
.gate footer { border-top: 1px solid var(--line); padding-top: 10px; }
|
|
a { color: var(--ink); }
|
|
.twocol {
|
|
display: grid;
|
|
grid-template-columns: minmax(0, 1fr) minmax(0, 1fr);
|
|
gap: 22px;
|
|
align-items: start;
|
|
}
|
|
.panel {
|
|
border: 1px solid var(--line);
|
|
border-radius: 8px;
|
|
padding: 18px;
|
|
background: #fff;
|
|
}
|
|
.panel p { color: var(--muted); }
|
|
.kv-grid {
|
|
display: grid;
|
|
grid-template-columns: repeat(2, minmax(0, 1fr));
|
|
gap: 10px 12px;
|
|
margin: 2px 0 0;
|
|
}
|
|
.kv-grid div {
|
|
min-width: 0;
|
|
padding: 9px 0 0;
|
|
border-top: 1px solid var(--line);
|
|
}
|
|
.kv-grid dt {
|
|
color: var(--muted);
|
|
font-size: 13px;
|
|
}
|
|
.kv-grid dd {
|
|
margin: 2px 0 0;
|
|
color: var(--ink);
|
|
font-size: 15px;
|
|
overflow-wrap: anywhere;
|
|
}
|
|
.issues, .evidence {
|
|
list-style: none;
|
|
padding: 0;
|
|
margin: 0;
|
|
display: grid;
|
|
gap: 12px;
|
|
}
|
|
.issues li, .evidence li {
|
|
border-left: 3px solid var(--line);
|
|
padding-left: 12px;
|
|
display: grid;
|
|
gap: 3px;
|
|
}
|
|
.muted { color: var(--muted); }
|
|
.actions-grid {
|
|
display: grid;
|
|
grid-template-columns: repeat(2, minmax(0, 1fr));
|
|
gap: 16px;
|
|
}
|
|
.annotations-grid {
|
|
display: grid;
|
|
grid-template-columns: repeat(2, minmax(0, 1fr));
|
|
gap: 16px;
|
|
}
|
|
.action-card {
|
|
border: 1px solid var(--line);
|
|
border-radius: 8px;
|
|
padding: 18px;
|
|
background: #fff;
|
|
display: grid;
|
|
gap: 10px;
|
|
min-width: 0;
|
|
}
|
|
.action-card.warn { border-left: 4px solid var(--warn); }
|
|
.action-card.block { border-left: 4px solid var(--block); }
|
|
.action-card.with-evidence {
|
|
grid-column: 1 / -1;
|
|
}
|
|
.action-evidence-panel {
|
|
display: grid;
|
|
gap: 12px;
|
|
border-top: 1px solid var(--line);
|
|
padding-top: 12px;
|
|
margin-top: 2px;
|
|
}
|
|
.action-evidence-panel h4 {
|
|
margin: 0;
|
|
color: var(--ink);
|
|
font-size: 16px;
|
|
font-weight: 500;
|
|
}
|
|
.action-evidence-panel p {
|
|
color: var(--muted);
|
|
font-size: 14px;
|
|
}
|
|
.action-evidence-grid {
|
|
display: grid;
|
|
grid-template-columns: repeat(2, minmax(0, 1fr));
|
|
gap: 14px;
|
|
}
|
|
.action-evidence-item {
|
|
border: 1px solid var(--line);
|
|
border-radius: 8px;
|
|
padding: 14px;
|
|
display: grid;
|
|
gap: 10px;
|
|
min-width: 0;
|
|
background: var(--soft);
|
|
}
|
|
.action-evidence-item.pending { border-left: 4px solid var(--warn); }
|
|
.action-evidence-item.accepted { border-left: 4px solid var(--pass); }
|
|
.action-evidence-item h4,
|
|
.action-evidence-item h5 {
|
|
margin: 0;
|
|
color: var(--ink);
|
|
font-weight: 500;
|
|
}
|
|
.action-evidence-item h4 { font-size: 16px; }
|
|
.action-evidence-item h5 { font-size: 14px; }
|
|
.action-evidence-item section {
|
|
border-top: 1px solid var(--line);
|
|
padding-top: 10px;
|
|
display: grid;
|
|
gap: 8px;
|
|
}
|
|
.action-command-details,
|
|
.action-runbook-details {
|
|
border-top: 1px solid var(--line);
|
|
padding-top: 10px;
|
|
display: grid;
|
|
gap: 8px;
|
|
}
|
|
.action-command-details summary,
|
|
.action-runbook-details summary {
|
|
cursor: pointer;
|
|
color: var(--ink);
|
|
font-size: 14px;
|
|
font-weight: 500;
|
|
}
|
|
.action-command-details[open] summary,
|
|
.action-runbook-details[open] summary {
|
|
margin-bottom: 8px;
|
|
}
|
|
.action-evidence-item dl {
|
|
display: grid;
|
|
grid-template-columns: 76px minmax(0, 1fr);
|
|
gap: 6px 10px;
|
|
margin: 0;
|
|
}
|
|
.action-evidence-item dt {
|
|
color: var(--ink);
|
|
font-size: 14px;
|
|
}
|
|
.action-evidence-item dd {
|
|
margin: 0;
|
|
min-width: 0;
|
|
}
|
|
.action-evidence-checks,
|
|
.action-command-list,
|
|
.action-runbook-list {
|
|
margin: 0;
|
|
display: grid;
|
|
gap: 8px;
|
|
}
|
|
.action-evidence-checks,
|
|
.action-command-list {
|
|
padding: 0;
|
|
list-style: none;
|
|
}
|
|
.action-runbook-list {
|
|
padding-left: 20px;
|
|
}
|
|
.action-evidence-check,
|
|
.action-command-list li {
|
|
display: grid;
|
|
gap: 3px;
|
|
min-width: 0;
|
|
border-top: 1px solid var(--line);
|
|
padding-top: 8px;
|
|
}
|
|
.action-evidence-check.blocked { border-left: 3px solid var(--warn); padding-left: 8px; }
|
|
.action-evidence-check.pass { border-left: 3px solid var(--pass); padding-left: 8px; }
|
|
.action-evidence-check span,
|
|
.action-command-list span {
|
|
color: var(--ink);
|
|
font-size: 13px;
|
|
}
|
|
.action-evidence-check code,
|
|
.action-evidence-check small,
|
|
.action-command-list code,
|
|
.action-runbook-list li {
|
|
color: var(--muted);
|
|
font-size: 13px;
|
|
overflow-wrap: anywhere;
|
|
}
|
|
.action-command-list code {
|
|
display: block;
|
|
white-space: normal;
|
|
}
|
|
.annotation-card {
|
|
border: 1px solid var(--line);
|
|
border-radius: 8px;
|
|
padding: 18px;
|
|
background: #fff;
|
|
display: grid;
|
|
gap: 10px;
|
|
min-width: 0;
|
|
}
|
|
.annotation-card.warning { border-left: 4px solid var(--warn); }
|
|
.annotation-card.blocker { border-left: 4px solid var(--block); }
|
|
.annotation-card.resolved { opacity: .72; }
|
|
.world-evidence-grid {
|
|
display: grid;
|
|
grid-template-columns: minmax(0, 1fr);
|
|
gap: 16px;
|
|
margin-top: 16px;
|
|
}
|
|
.world-evidence-card {
|
|
border: 1px solid var(--line);
|
|
border-radius: 8px;
|
|
padding: 18px;
|
|
background: #fff;
|
|
display: grid;
|
|
gap: 12px;
|
|
min-width: 0;
|
|
}
|
|
.world-evidence-card.pending { border-left: 4px solid var(--warn); }
|
|
.world-evidence-card.accepted { border-left: 4px solid var(--pass); }
|
|
.world-evidence-card span,
|
|
.world-evidence-card p,
|
|
.world-evidence-card dd,
|
|
.world-evidence-card li {
|
|
color: var(--muted);
|
|
font-size: 14px;
|
|
overflow-wrap: anywhere;
|
|
}
|
|
.world-evidence-card dl {
|
|
display: grid;
|
|
grid-template-columns: 86px minmax(0, 1fr);
|
|
gap: 6px 10px;
|
|
margin: 0;
|
|
}
|
|
.world-evidence-card dt { color: var(--ink); font-size: 14px; }
|
|
.world-evidence-card dd { margin: 0; }
|
|
.world-evidence-columns {
|
|
display: grid;
|
|
grid-template-columns: repeat(3, minmax(0, 1fr));
|
|
gap: 12px;
|
|
border-top: 1px solid var(--line);
|
|
padding-top: 12px;
|
|
}
|
|
.world-evidence-columns h4 {
|
|
margin: 0 0 6px;
|
|
color: var(--ink);
|
|
font-size: 14px;
|
|
font-weight: 500;
|
|
}
|
|
.world-evidence-columns ul {
|
|
margin: 0;
|
|
padding-left: 18px;
|
|
}
|
|
.world-source-panel {
|
|
border-top: 1px solid var(--line);
|
|
padding-top: 12px;
|
|
}
|
|
.world-runbook-panel {
|
|
border-top: 1px solid var(--line);
|
|
padding-top: 12px;
|
|
}
|
|
.world-runbook-panel h4,
|
|
.world-source-panel h4 {
|
|
margin: 0 0 8px;
|
|
color: var(--ink);
|
|
font-size: 14px;
|
|
font-weight: 500;
|
|
}
|
|
.world-runbook-panel ul {
|
|
margin: 0;
|
|
padding-left: 18px;
|
|
}
|
|
.world-source-checks {
|
|
list-style: none;
|
|
padding: 0;
|
|
margin: 0;
|
|
display: grid;
|
|
gap: 8px;
|
|
}
|
|
.world-source-check {
|
|
display: grid;
|
|
grid-template-columns: minmax(0, .7fr) minmax(0, .6fr) minmax(0, 1fr);
|
|
gap: 10px;
|
|
align-items: start;
|
|
min-width: 0;
|
|
padding: 9px 0 0;
|
|
border-top: 1px solid var(--line);
|
|
}
|
|
.world-source-check span {
|
|
color: var(--ink);
|
|
font-size: 14px;
|
|
}
|
|
.world-source-check code,
|
|
.world-source-check small {
|
|
color: var(--muted);
|
|
font-size: 13px;
|
|
overflow-wrap: anywhere;
|
|
}
|
|
.world-source-check.pass { border-left: 3px solid var(--pass); padding-left: 8px; }
|
|
.world-source-check.blocked { border-left: 3px solid var(--warn); padding-left: 8px; }
|
|
.output-review-grid {
|
|
display: grid;
|
|
grid-template-columns: minmax(0, 1fr);
|
|
gap: 16px;
|
|
margin-top: 16px;
|
|
}
|
|
.output-review-card {
|
|
border: 1px solid var(--line);
|
|
border-radius: 8px;
|
|
padding: 18px;
|
|
background: #fff;
|
|
display: grid;
|
|
gap: 12px;
|
|
min-width: 0;
|
|
}
|
|
.output-review-card.awaiting-decision,
|
|
.output-review-card.fix-decision { border-left: 4px solid var(--warn); }
|
|
.output-review-card.adjudicated { border-left: 4px solid var(--pass); }
|
|
.output-review-card span,
|
|
.output-review-card p,
|
|
.output-review-card dd,
|
|
.output-review-card li {
|
|
color: var(--muted);
|
|
font-size: 14px;
|
|
overflow-wrap: anywhere;
|
|
}
|
|
.output-review-card dl {
|
|
display: grid;
|
|
grid-template-columns: 86px minmax(0, 1fr);
|
|
gap: 6px 10px;
|
|
margin: 0;
|
|
}
|
|
.output-review-card dt { color: var(--ink); font-size: 14px; }
|
|
.output-review-card dd { margin: 0; min-width: 0; }
|
|
.output-review-steps {
|
|
display: grid;
|
|
grid-template-columns: repeat(3, minmax(0, 1fr));
|
|
gap: 12px;
|
|
border-top: 1px solid var(--line);
|
|
padding-top: 12px;
|
|
}
|
|
.output-review-steps h4 {
|
|
margin: 0 0 6px;
|
|
color: var(--ink);
|
|
font-size: 14px;
|
|
font-weight: 500;
|
|
}
|
|
.output-review-steps ul {
|
|
margin: 0;
|
|
padding-left: 18px;
|
|
}
|
|
.output-review-commands,
|
|
.output-review-fields {
|
|
list-style: none;
|
|
padding: 0;
|
|
display: grid;
|
|
gap: 8px;
|
|
}
|
|
.output-review-commands li,
|
|
.output-review-fields li {
|
|
display: grid;
|
|
gap: 2px;
|
|
min-width: 0;
|
|
padding: 8px 0 0;
|
|
border-top: 1px solid var(--line);
|
|
}
|
|
.output-review-commands code {
|
|
overflow-wrap: anywhere;
|
|
white-space: normal;
|
|
display: block;
|
|
}
|
|
.world-intake-grid {
|
|
display: grid;
|
|
grid-template-columns: minmax(0, 1fr);
|
|
gap: 16px;
|
|
margin-top: 16px;
|
|
}
|
|
.world-intake-card {
|
|
border: 1px solid var(--line);
|
|
border-radius: 8px;
|
|
padding: 18px;
|
|
background: #fff;
|
|
display: grid;
|
|
gap: 12px;
|
|
min-width: 0;
|
|
}
|
|
.world-intake-card.awaiting-submission,
|
|
.world-intake-card.fix-submission,
|
|
.world-intake-card.fix-template { border-left: 4px solid var(--warn); }
|
|
.world-intake-card.ready-for-ledger-review,
|
|
.world-intake-card.accepted { border-left: 4px solid var(--pass); }
|
|
.world-intake-card span,
|
|
.world-intake-card p,
|
|
.world-intake-card dd,
|
|
.world-intake-card li {
|
|
color: var(--muted);
|
|
font-size: 14px;
|
|
overflow-wrap: anywhere;
|
|
}
|
|
.world-intake-card dl {
|
|
display: grid;
|
|
grid-template-columns: 86px minmax(0, 1fr);
|
|
gap: 6px 10px;
|
|
margin: 0;
|
|
}
|
|
.world-intake-card dt { color: var(--ink); font-size: 14px; }
|
|
.world-intake-card dd { margin: 0; min-width: 0; }
|
|
.world-intake-steps {
|
|
display: grid;
|
|
grid-template-columns: repeat(2, minmax(0, 1fr));
|
|
gap: 12px;
|
|
border-top: 1px solid var(--line);
|
|
padding-top: 12px;
|
|
}
|
|
.world-intake-steps h4 {
|
|
margin: 0 0 6px;
|
|
color: var(--ink);
|
|
font-size: 14px;
|
|
font-weight: 500;
|
|
}
|
|
.world-intake-steps ul {
|
|
margin: 0;
|
|
padding-left: 18px;
|
|
}
|
|
.world-intake-commands {
|
|
list-style: none;
|
|
padding: 0;
|
|
display: grid;
|
|
gap: 8px;
|
|
}
|
|
.world-intake-commands li {
|
|
display: grid;
|
|
gap: 2px;
|
|
min-width: 0;
|
|
padding: 8px 0 0;
|
|
border-top: 1px solid var(--line);
|
|
}
|
|
.world-intake-commands code {
|
|
overflow-wrap: anywhere;
|
|
white-space: normal;
|
|
display: block;
|
|
}
|
|
.section-subtitle {
|
|
margin-top: 22px;
|
|
margin-bottom: 12px;
|
|
color: var(--ink);
|
|
}
|
|
.waiver-summary { margin-top: 16px; }
|
|
.waiver-candidate-grid {
|
|
display: grid;
|
|
grid-template-columns: minmax(0, 1fr);
|
|
gap: 16px;
|
|
}
|
|
.waiver-card {
|
|
border: 1px solid var(--line);
|
|
border-radius: 8px;
|
|
padding: 18px;
|
|
background: #fff;
|
|
display: grid;
|
|
gap: 12px;
|
|
min-width: 0;
|
|
}
|
|
.waiver-card.allowed { border-left: 4px solid var(--warn); }
|
|
.waiver-card.blocked { border-left: 4px solid var(--block); }
|
|
.waiver-card span,
|
|
.waiver-card p,
|
|
.waiver-card dd,
|
|
.waiver-card li {
|
|
color: var(--muted);
|
|
font-size: 14px;
|
|
overflow-wrap: anywhere;
|
|
}
|
|
.waiver-card dl {
|
|
display: grid;
|
|
grid-template-columns: 86px minmax(0, 1fr);
|
|
gap: 6px 10px;
|
|
margin: 0;
|
|
}
|
|
.waiver-card dt {
|
|
color: var(--ink);
|
|
font-size: 14px;
|
|
}
|
|
.waiver-card dd { margin: 0; min-width: 0; }
|
|
.waiver-review {
|
|
border-top: 1px solid var(--line);
|
|
padding-top: 12px;
|
|
}
|
|
.waiver-review h4 {
|
|
margin: 0 0 6px;
|
|
color: var(--ink);
|
|
font-size: 14px;
|
|
font-weight: 500;
|
|
}
|
|
.waiver-review ul {
|
|
margin: 0;
|
|
padding-left: 18px;
|
|
}
|
|
.waiver-command {
|
|
display: grid;
|
|
gap: 4px;
|
|
border-top: 1px solid var(--line);
|
|
padding-top: 12px;
|
|
}
|
|
.waiver-command span {
|
|
color: var(--ink);
|
|
font-size: 14px;
|
|
}
|
|
.waiver-command code {
|
|
display: block;
|
|
max-width: 100%;
|
|
overflow-wrap: anywhere;
|
|
white-space: normal;
|
|
}
|
|
.action-card span,
|
|
.annotation-card span,
|
|
.action-card small,
|
|
.annotation-card small,
|
|
.action-card footer,
|
|
.annotation-card footer,
|
|
.action-card dd {
|
|
color: var(--muted);
|
|
font-size: 14px;
|
|
overflow-wrap: anywhere;
|
|
}
|
|
.annotation-card dd { color: var(--muted); font-size: 14px; overflow-wrap: anywhere; }
|
|
.action-card dl, .annotation-card dl {
|
|
display: grid;
|
|
grid-template-columns: 80px minmax(0, 1fr);
|
|
gap: 6px 10px;
|
|
margin: 0;
|
|
}
|
|
.action-card dt, .annotation-card dt { color: var(--ink); font-size: 14px; }
|
|
.action-card dd, .annotation-card dd { margin: 0; }
|
|
.source-ref-list {
|
|
list-style: none;
|
|
padding: 0;
|
|
margin: 0;
|
|
display: grid;
|
|
gap: 6px;
|
|
}
|
|
.source-ref-list li {
|
|
display: grid;
|
|
gap: 2px;
|
|
min-width: 0;
|
|
padding: 8px 0 0;
|
|
border-top: 1px solid var(--line);
|
|
}
|
|
.source-ref-list a,
|
|
.source-ref-list span {
|
|
overflow-wrap: anywhere;
|
|
font-family: ui-monospace, SFMono-Regular, Menlo, Consolas, monospace;
|
|
font-size: 13px;
|
|
}
|
|
.source-ref-list small {
|
|
font-size: 12px;
|
|
color: var(--muted);
|
|
}
|
|
code {
|
|
font-family: ui-monospace, SFMono-Regular, Menlo, Consolas, monospace;
|
|
font-size: 13px;
|
|
}
|
|
@media (max-width: 980px) {
|
|
.metrics, .gates, .twocol, .actions-grid, .annotations-grid, .action-evidence-grid, .action-evidence-item dl, .output-review-grid, .output-review-steps, .world-evidence-grid, .world-evidence-columns, .world-source-check, .world-intake-grid, .world-intake-steps, .waiver-candidate-grid, .waiver-card dl, .kv-grid { grid-template-columns: 1fr; }
|
|
main { padding: 32px 18px 60px; }
|
|
nav { justify-content: flex-start; overflow-x: auto; flex-wrap: nowrap; }
|
|
nav a { flex: 0 0 auto; }
|
|
}
|
|
</style>
|
|
</head>
|
|
<body>
|
|
<nav><a href='#overview'>审查总览</a><a href='#intent'>意图画布</a><a href='#trigger'>触发实验</a><a href='#output'>输出实验</a><a href='#actions'>修复动作</a><a href='#annotations'>审查批注</a><a href='#runtime'>运行矩阵</a><a href='#trust'>信任报告</a><a href='#permissions'>权限批准</a><a href='#permission-probes'>权限探针</a><a href='#atlas'>组合治理</a><a href='#telemetry'>运营回路</a><a href='#waivers'>人工批准</a><a href='#world-class'>世界证据</a><a href='#registry'>注册审计</a><a href='#release'>发布路线</a></nav>
|
|
<main>
|
|
<header id="overview">
|
|
<div class="eyebrow">Review Studio 2.0</div>
|
|
<h1>Yao Meta Skill</h1>
|
|
<p class="lede">Create, refactor, evaluate, and package agent skills from workflows, prompts, transcripts, docs, or notes. Use when asked to create a skill, turn a repeated process into a reusable skill, improve an existing skill, add evals, or package a skill for team reuse.</p>
|
|
<div class="decision">
|
|
<span>审查结论</span>
|
|
<strong>review</strong>
|
|
<span>Score 91/100</span>
|
|
</div>
|
|
</header>
|
|
|
|
<section>
|
|
<h2>核心指标</h2>
|
|
<div class="metrics"><article class='metric'><span>Skill IR</span><strong>2.0.0</strong><p>5 targets in platform-neutral contract</p></article><article class='metric'><span>Compiler</span><strong>5/5</strong><p>target contracts compiled from Skill IR</p></article><article class='metric'><span>Output Delta</span><strong>100.0</strong><p>5 cases; 1 file-backed</p></article><article class='metric'><span>Exec Runs</span><strong>10</strong><p>command 10; model 0; recorded 0</p></article><article class='metric'><span>Blind A/B</span><strong>5</strong><p>review pairs hide baseline vs with-skill labels</p></article><article class='metric'><span>Review Kit</span><strong>0/5</strong><p>pending 5; answer key hidden</p></article><article class='metric'><span>Review A/B</span><strong>0/5</strong><p>adjudication decisions; pending 5</p></article><article class='metric'><span>Public Claim</span><strong>blocked</strong><p>5 blockers; local reproducible true</p></article><article class='metric'><span>Blueprint</span><strong>20/20</strong><p>2.0 coverage; extensions partial 1, planned 0; evidence pending 4</p></article><article class='metric'><span>Runtime</span><strong>5/5</strong><p>target conformance pass rate</p></article><article class='metric'><span>Perm Probe</span><strong>4/4</strong><p>0 native; 4 installer-enforced</p></article><article class='metric'><span>Trust</span><strong>0</strong><p>106 scripts scanned; secrets found</p></article><article class='metric'><span>Py Compat</span><strong>0</strong><p>171 files scanned for Python 3.11</p></article><article class='metric'><span>Arch Debt</span><strong>0</strong><p>886 largest lines; 34 CLI handlers</p></article><article class='metric'><span>Atlas</span><strong>5</strong><p>12 scanned skills; route collisions</p></article><article class='metric'><span>Drift</span><strong>low</strong><p>1 metadata events; 0 missed triggers</p></article><article class='metric'><span>Waivers</span><strong>0</strong><p>0 gates covered; human risk decisions</p></article><article class='metric'><span>Intake</span><strong>4/4</strong><p>0 valid submissions; 0 invalid</p></article><article class='metric'><span>Claim Guard</span><strong>0</strong><p>76 public surfaces scanned</p></article><article class='metric'><span>Notes</span><strong>0/0</strong><p>0 open blocker annotations</p></article><article class='metric'><span>Registry</span><strong>1.1.0</strong><p>5 targets; MIT license</p></article><article class='metric'><span>Archive</span><strong>pass</strong><p>586 zip entries; package verification</p></article><article class='metric'><span>Install</span><strong>pass</strong><p>4 adapters; 12 permissions enforced; 0 permission failures</p></article><article class='metric'><span>Upgrade</span><strong>minor</strong><p>declared minor; 0 breaking changes</p></article></div>
|
|
</section>
|
|
|
|
<section>
|
|
<h2>审查闸门</h2>
|
|
<div class="gates"><article class='gate pass'><div><span>通过</span><h3>意图画布</h3></div><p>intent confidence 100/100; Intent is clear enough to package the first routeable version.</p><footer>reports/intent-confidence.json <a href='intent-confidence.md'>证据</a></footer></article><article class='gate pass'><div><span>通过</span><h3>触发实验</h3></div><p>13 trigger cases; 0 misroutes; 0 ambiguous</p><footer>reports/route_scorecard.json <a href='route_scorecard.md'>证据</a></footer></article><article class='gate warn'><div><span>关注</span><h3>输出实验</h3></div><p>5/5 cases; with-skill 100.0; baseline 0.0; file-backed 1; near-neighbor 1; blind A/B 5; exec 10; command 10; model 0; recorded 0; reviewed 0/5; review pending 5</p><footer>reports/output_quality_scorecard.json <a href='output_quality_scorecard.md'>证据</a></footer></article><article class='gate pass'><div><span>通过</span><h3>上下文</h3></div><p>initial load 960/1000; deferred 416415/120000; top deferred scripts 367825; resource governance governed; quality density 135.4</p><footer>reports/context_budget.json <a href='context_budget.md'>证据</a></footer></article><article class='gate pass'><div><span>通过</span><h3>运行矩阵</h3></div><p>5 / 5 targets pass</p><footer>reports/conformance_matrix.json <a href='conformance_matrix.md'>证据</a></footer></article><article class='gate pass'><div><span>通过</span><h3>信任报告</h3></div><p>0 secrets; 106 scripts; 3 network-capable scripts; 0 help smoke failures</p><footer>reports/security_trust_report.json <a href='security_trust_report.md'>证据</a></footer></article><article class='gate pass'><div><span>通过</span><h3>Python 兼容</h3></div><p>Python 3.11; 171 files; 0 compatibility issues; 0 syntax; 0 f-string 3.11 hazards</p><footer>reports/python_compatibility.json <a href='python_compatibility.md'>证据</a></footer></article><article class='gate pass'><div><span>通过</span><h3>架构维护</h3></div><p>168 Python files; 0 hotspots; 0 blockers; largest 886 lines; 34 CLI handlers</p><footer>reports/architecture_maintainability.json <a href='architecture_maintainability.md'>证据</a></footer></article><article class='gate pass'><div><span>通过</span><h3>权限批准</h3></div><p>3/3 permissions approved; gaps 0; required file_write, network, subprocess</p><footer>reports/security_trust_report.json + security/permission_policy.json <a href='../security/permission_policy.md'>证据</a></footer></article><article class='gate pass'><div><span>通过</span><h3>权限探针</h3></div><p>4/4 targets probed; native 0; metadata fallback 4; installer 4; residual risks 4</p><footer>reports/runtime_permission_probes.json <a href='runtime_permission_probes.md'>证据</a></footer></article><article class='gate pass'><div><span>通过</span><h3>组合治理</h3></div><p>12 skills, 1 actionable; 0 actionable route collisions; 0 actionable owner gaps; 0 actionable stale; 0 actionable drift; 24 scoped non-actionable issues</p><footer>reports/skill_atlas.json <a href='skill_atlas.html'>证据</a></footer></article><article class='gate pass'><div><span>通过</span><h3>运营回路</h3></div><p>1 metadata events; adoption 100.0; missed 0; bad-output 0; risk low</p><footer>reports/adoption_drift_report.json <a href='adoption_drift_report.md'>证据</a></footer></article><article class='gate warn'><div><span>关注</span><h3>人工批准</h3></div><p>0 active waivers; 1 warning gates still need reviewer decision</p><footer>reports/review_waivers.json <a href='review_waivers.md'>证据</a></footer></article><article class='gate warn'><div><span>关注</span><h3>世界证据</h3></div><p>4 pending world-class evidence entries; 1 human pending; 3 external pending; source checks 7/13 pass; 6 blocked; overclaim guard true</p><footer>reports/world_class_evidence_ledger.json <a href='world_class_evidence_ledger.md'>证据</a></footer></article><article class='gate pass'><div><span>通过</span><h3>注册审计</h3></div><p>yao-meta-skill 1.1.0; 6/6 compatibility entries pass; install pass with 4 adapters; installer permissions 12 enforced / 0 failures</p><footer>reports/registry_audit.json + reports/install_simulation.json <a href='registry_audit.md'>证据</a></footer></article><article class='gate pass'><div><span>通过</span><h3>发布路线</h3></div><p>0 promote; 3 keep current; 0 blocked; upgrade minor declared / minor recommended</p><footer>reports/promotion_decisions.json + reports/upgrade_check.json + docs/migration-v2.md <a href='promotion_decisions.md'>证据</a></footer></article></div>
|
|
</section>
|
|
|
|
<div class="twocol">
|
|
<section><h2>阻断事项</h2><p class='muted'>无。</p></section>
|
|
<section><h2>关注事项</h2><ul class='issues'><li><strong>输出实验</strong><span>5/5 cases; with-skill 100.0; baseline 0.0; file-backed 1; near-neighbor 1; blind A/B 5; exec 10; command 10; model 0; recorded 0; reviewed 0/5; review pending 5</span></li><li><strong>人工批准</strong><span>0 active waivers; 1 warning gates still need reviewer decision</span></li><li><strong>世界证据</strong><span>4 pending world-class evidence entries; 1 human pending; 3 external pending; source checks 7/13 pass; 6 blocked; overclaim guard true</span></li></ul></section>
|
|
</div>
|
|
|
|
<section id="actions">
|
|
<h2>修复动作</h2>
|
|
<div class="actions-grid"><article class='action-card warn'><div><span>关注</span><h3>输出实验</h3></div><p>补足 output eval 覆盖、execution evidence、blind A/B 和 reviewer adjudication。</p><small>没有输出质量和人工盲评证据时，Skill 只能证明会触发，不能证明输出真的更好且经得起审查。</small><dl><dt>修复位置</dt><dd>evals/output/cases.jsonl + reports/output_quality_scorecard.md + reports/output_review_kit.html + reports/output_review_adjudication.md</dd><dt>验证命令</dt><dd><code>python3 scripts/adjudicate_output_review.py --write-template && python3 scripts/yao.py output-review</code></dd></dl><ul class='source-ref-list'><li><a href='../evals/output/cases.jsonl'>evals/output/cases.jsonl:1</a><small>output eval cases · eval</small></li><li><a href='output_quality_scorecard.md'>reports/output_quality_scorecard.md:1</a><small>output scorecard · report</small></li><li><a href='output_execution_runs.md'>reports/output_execution_runs.md:1</a><small>output execution runs · report</small></li><li><a href='output_blind_review_pack.md'>reports/output_blind_review_pack.md:1</a><small>blind A/B review pack · report</small></li><li><a href='output_review_kit.html'>reports/output_review_kit.html:6</a><small>reviewer cockpit · report</small></li><li><a href='output_review_adjudication.md'>reports/output_review_adjudication.md:1</a><small>review adjudication · report</small></li></ul><footer>reports/output_quality_scorecard.json <a href='output_quality_scorecard.md'>打开证据</a></footer></article><article class='action-card warn'><div><span>关注</span><h3>人工批准</h3></div><p>对保留的 warning 写入 reviewer、理由、范围和到期时间，或修掉 warning。</p><small>warning 可以被接受，但必须可审计、会过期，并且不能掩盖 blocker。</small><dl><dt>修复位置</dt><dd>reports/review_waivers.md</dd><dt>验证命令</dt><dd><code>python3 scripts/render_review_waivers.py .</code></dd></dl><ul class='source-ref-list'><li><a href='review_waivers.md'>reports/review_waivers.md:1</a><small>waiver ledger · report</small></li><li><a href='../references/review-waiver-method.md'>references/review-waiver-method.md:1</a><small>waiver method · method</small></li></ul><footer>reports/review_waivers.json <a href='review_waivers.md'>打开证据</a></footer></article><article class='action-card warn with-evidence'><div><span>关注</span><h3>世界证据</h3></div><p>补齐 provider、真人盲评、原生权限执行和真实客户端遥测证据，或明确本次发布不声明 world-class 完成。</p><small>世界级结论必须来自已接受的外部/人工证据；计划、metadata fallback、待评审和本地命令都不能替代完成证据。</small><dl><dt>修复位置</dt><dd>reports/world_class_operator_runbook.html + reports/world_class_evidence_ledger.md + reports/world_class_evidence_intake.md + reports/world_class_submission_review.md</dd><dt>验证命令</dt><dd><code>python3 scripts/yao.py world-class-runbook . --submissions-dir evidence/world_class/submissions && python3 scripts/yao.py world-class-ledger . --submissions-dir evidence/world_class/submissions && python3 scripts/yao.py review-studio .</code></dd></dl><section class='action-evidence-panel'><h4>证据采集</h4><p>以下条目仍需真实外部或人工证据；提交文件、校验命令和阻断检查必须同时闭环。</p><div class='action-evidence-grid'><article class='action-evidence-item pending'><div><span>pending · external</span><h4>Provider Holdout</h4></div><p>model-executed 0; token-observed 0</p><dl><dt>提交</dt><dd><code>evidence/world_class/submissions/provider-holdout.json</code></dd><dt>模板</dt><dd><code>evidence/world_class/templates/provider-holdout.intake.json</code></dd><dt>阻断</dt><dd>2 blocked / 1 pass</dd><dt>下一步</dt><dd>Run provider-backed holdout cases with real credentials and commit only aggregate evidence.</dd></dl><section><h5>阻断检查</h5><ul class='action-evidence-checks'><li class='action-evidence-check blocked'><span>Provider model run</span><code>model_executed_count: 0 / >0</code><small>Run provider-backed output-exec with real credentials.</small></li><li class='action-evidence-check blocked'><span>Token usage observed</span><code>token_observed_count: 0 / >0</code><small>Provider execution should return non-estimated token usage.</small></li></ul></section><details class='action-command-details'><summary>操作命令</summary><ul class='action-command-list'><li><span>准备提交</span><code>python3 scripts/yao.py world-class-submission-kit . --evidence-key provider-holdout --output-dir evidence/world_class/submissions</code></li><li><span>校验入口</span><code>python3 scripts/yao.py world-class-intake . --submissions-dir evidence/world_class/submissions</code></li><li><span>审查提交</span><code>python3 scripts/yao.py world-class-submission-review . --submissions-dir evidence/world_class/submissions</code></li><li><span>刷新台账</span><code>python3 scripts/yao.py world-class-ledger . --submissions-dir evidence/world_class/submissions</code></li></ul></details><details class='action-runbook-details'><summary>首要步骤</summary><ol class='action-runbook-list'><li>YAO_OUTPUT_EVAL_MODEL=gpt-4.1-mini OPENAI_API_KEY=<redacted> python3 scripts/yao.py output-exec --provider-runner openai --timeout-seconds 60</li><li>python3 scripts/yao.py skill-os2-audit . --generated-at <YYYY-MM-DD></li><li>Copy evidence/world_class/templates/provider-holdout.intake.json to evidence/world_class/submissions/provider-holdout.json and fill only real evidence fields.</li></ol></details></article><article class='action-evidence-item pending'><div><span>pending · human</span><h4>Human Adjudication</h4></div><p>0/5 decisions; pending 5</p><dl><dt>提交</dt><dd><code>evidence/world_class/submissions/human-adjudication.json</code></dd><dt>模板</dt><dd><code>evidence/world_class/templates/human-adjudication.intake.json</code></dd><dt>阻断</dt><dd>2 blocked / 2 pass</dd><dt>下一步</dt><dd>Record real A/B choices in the decision template, then regenerate adjudication.</dd></dl><section><h5>阻断检查</h5><ul class='action-evidence-checks'><li class='action-evidence-check blocked'><span>No pending decisions</span><code>pending_count: 5 / ==0</code><small>Record a reviewer choice for every pair.</small></li><li class='action-evidence-check blocked'><span>Judgments complete</span><code>judgment_count: 0 / ==pair_count</code><small>Every pair needs one valid human judgment.</small></li></ul></section><details class='action-command-details'><summary>操作命令</summary><ul class='action-command-list'><li><span>准备提交</span><code>python3 scripts/yao.py world-class-submission-kit . --evidence-key human-adjudication --output-dir evidence/world_class/submissions</code></li><li><span>校验入口</span><code>python3 scripts/yao.py world-class-intake . --submissions-dir evidence/world_class/submissions</code></li><li><span>审查提交</span><code>python3 scripts/yao.py world-class-submission-review . --submissions-dir evidence/world_class/submissions</code></li><li><span>刷新台账</span><code>python3 scripts/yao.py world-class-ledger . --submissions-dir evidence/world_class/submissions</code></li></ul></details><details class='action-runbook-details'><summary>首要步骤</summary><ol class='action-runbook-list'><li>python3 scripts/yao.py output-review-kit --write-template</li><li>Open reports/output_review_kit.md and choose A or B for each pair without opening the answer key.</li><li>python3 scripts/adjudicate_output_review.py --write-template</li></ol></details></article><article class='action-evidence-item pending'><div><span>pending · external</span><h4>Native Permission Enforcement</h4></div><p>native-enforced targets 0; installer-enforced targets 4</p><dl><dt>提交</dt><dd><code>evidence/world_class/submissions/native-permission-enforcement.json</code></dd><dt>模板</dt><dd><code>evidence/world_class/templates/native-permission-enforcement.intake.json</code></dd><dt>阻断</dt><dd>1 blocked / 2 pass</dd><dt>下一步</dt><dd>Integrate a real target-client or external installer runtime guard before claiming native permission enforcement.</dd></dl><section><h5>阻断检查</h5><ul class='action-evidence-checks'><li class='action-evidence-check blocked'><span>Native enforcement</span><code>native_enforcement_count: 0 / >0</code><small>Collect real target-client or external runtime guard proof.</small></li></ul></section><details class='action-command-details'><summary>操作命令</summary><ul class='action-command-list'><li><span>准备提交</span><code>python3 scripts/yao.py world-class-submission-kit . --evidence-key native-permission-enforcement --output-dir evidence/world_class/submissions</code></li><li><span>校验入口</span><code>python3 scripts/yao.py world-class-intake . --submissions-dir evidence/world_class/submissions</code></li><li><span>审查提交</span><code>python3 scripts/yao.py world-class-submission-review . --submissions-dir evidence/world_class/submissions</code></li><li><span>刷新台账</span><code>python3 scripts/yao.py world-class-ledger . --submissions-dir evidence/world_class/submissions</code></li></ul></details><details class='action-runbook-details'><summary>首要步骤</summary><ol class='action-runbook-list'><li>Implement or connect a real target client or external installer runtime guard that blocks undeclared network, file_write, or subprocess capabilities.</li><li>Update the generated target adapter only when the guard is actually enforced by that target.</li><li>python3 scripts/yao.py package . --platform openai --platform claude --platform generic --platform vscode --output-dir dist --zip</li></ol></details></article><article class='action-evidence-item pending'><div><span>pending · external</span><h4>Native Client Telemetry</h4></div><p>external source events 0; adoption samples 1</p><dl><dt>提交</dt><dd><code>evidence/world_class/submissions/native-client-telemetry.json</code></dd><dt>模板</dt><dd><code>evidence/world_class/templates/native-client-telemetry.intake.json</code></dd><dt>阻断</dt><dd>1 blocked / 2 pass</dd><dt>下一步</dt><dd>Install a real client against the native host and import production metadata-only events.</dd></dl><section><h5>阻断检查</h5><ul class='action-evidence-checks'><li class='action-evidence-check blocked'><span>External events</span><code>external_source_events: 0 / >0</code><small>Import at least one metadata-only event from a real client.</small></li></ul></section><details class='action-command-details'><summary>操作命令</summary><ul class='action-command-list'><li><span>准备提交</span><code>python3 scripts/yao.py world-class-submission-kit . --evidence-key native-client-telemetry --output-dir evidence/world_class/submissions</code></li><li><span>校验入口</span><code>python3 scripts/yao.py world-class-intake . --submissions-dir evidence/world_class/submissions</code></li><li><span>审查提交</span><code>python3 scripts/yao.py world-class-submission-review . --submissions-dir evidence/world_class/submissions</code></li><li><span>刷新台账</span><code>python3 scripts/yao.py world-class-ledger . --submissions-dir evidence/world_class/submissions</code></li></ul></details><details class='action-runbook-details'><summary>首要步骤</summary><ol class='action-runbook-list'><li>python3 scripts/telemetry_native_host.py . --write-launcher /tmp/yao-telemetry-host.sh --write-manifest /tmp/yao-telemetry-host.json --allowed-origin chrome-extension://<extension-id>/</li><li>Install the generated native messaging manifest for the real client and send at least one accepted skill_activation or skill_output event.</li><li>python3 scripts/yao.py telemetry-import . --input-jsonl .yao/telemetry_spool/external_events.jsonl</li></ol></details></article></div></section><ul class='source-ref-list'><li><a href='world_class_evidence_ledger.md'>reports/world_class_evidence_ledger.md:1</a><small>world-class evidence ledger · report</small></li><li><a href='world_class_evidence_plan.md'>reports/world_class_evidence_plan.md:1</a><small>world-class evidence plan · report</small></li><li><a href='world_class_evidence_intake.md'>reports/world_class_evidence_intake.md:1</a><small>world-class evidence intake · report</small></li><li><a href='world_class_submission_review.md'>reports/world_class_submission_review.md:1</a><small>world-class submission review · report</small></li><li><a href='world_class_claim_guard.md'>reports/world_class_claim_guard.md:1</a><small>world-class claim guard · report</small></li><li><a href='../evidence/world_class/intake.schema.json'>evidence/world_class/intake.schema.json:4</a><small>evidence intake schema · schema</small></li><li><a href='../evidence/world_class/templates/provider-holdout.intake.json'>evidence/world_class/templates/provider-holdout.intake.json:3</a><small>provider intake template · template</small></li><li><a href='../evidence/world_class/templates/human-adjudication.intake.json'>evidence/world_class/templates/human-adjudication.intake.json:3</a><small>human intake template · template</small></li><li><a href='../evidence/world_class/templates/native-permission-enforcement.intake.json'>evidence/world_class/templates/native-permission-enforcement.intake.json:3</a><small>permission intake template · template</small></li><li><a href='../evidence/world_class/templates/native-client-telemetry.intake.json'>evidence/world_class/templates/native-client-telemetry.intake.json:3</a><small>telemetry intake template · template</small></li><li><a href='skill_os2_audit.md'>reports/skill_os2_audit.md:1</a><small>Skill OS 2.0 audit · report</small></li><li><a href='output_review_decisions.json'>reports/output_review_decisions.json:6</a><small>human review decisions · report</small></li><li><a href='runtime_permission_probes.md'>reports/runtime_permission_probes.md:1</a><small>runtime permission probes · report</small></li><li><a href='adoption_drift_report.md'>reports/adoption_drift_report.md:1</a><small>adoption drift · report</small></li></ul><footer>reports/world_class_evidence_ledger.json <a href='world_class_evidence_ledger.md'>打开证据</a></footer></article></div>
|
|
</section>
|
|
|
|
<section id="annotations">
|
|
<h2>审查批注</h2>
|
|
<p class="muted">当前批注：0 条批注；开放 0；阻断 0</p>
|
|
<div class="annotations-grid"><p class='muted'>当前没有 reviewer 批注。</p></div>
|
|
</section>
|
|
|
|
<section id="intent" class="twocol">
|
|
<div class="panel">
|
|
<h2>意图画布</h2>
|
|
<p>Turn repeated workflows, prompts, transcripts, runbooks, documents, or existing skill packages into routeable, evaluable, packageable, and governable agent skills for personal, team, library, or governed reuse. Primary output: A working skill package with lean SKILL.md, aligned agents/interface.yaml, justified references, scripts only when useful, eval evidence, reports, packaging metadata, and clear next iteration recommendations.. Exclusions: one-off writing, translation, explanation, or brainstorming requests that do not need a reusable skill, general code review or debugging unless the user is packaging that workflow as a skill, raw private material that was not intentionally supplied as skill evidence, platform-specific plugin creation when the request is not about an agent skill package.</p>
|
|
</div>
|
|
<div class="panel">
|
|
<h2>证据路径</h2>
|
|
<ul class="evidence"><li><strong>skill_overview</strong><span>reports/skill-overview.html</span></li><li><strong>review_viewer</strong><span>reports/review-viewer.html</span></li><li><strong>output_eval</strong><span>reports/output_quality_scorecard.md</span></li><li><strong>output_execution</strong><span>reports/output_execution_runs.md</span></li><li><strong>output_blind_review</strong><span>reports/output_blind_review_pack.md</span></li><li><strong>output_review_kit</strong><span>reports/output_review_kit.md</span></li><li><strong>output_review_kit_html</strong><span>reports/output_review_kit.html</span></li><li><strong>output_review_decisions</strong><span>reports/output_review_decisions.json</span></li><li><strong>output_review_adjudication</strong><span>reports/output_review_adjudication.md</span></li><li><strong>benchmark_reproducibility</strong><span>reports/benchmark_reproducibility.md</span></li><li><strong>skill_os2_coverage</strong><span>reports/skill_os2_coverage.md</span></li><li><strong>runtime_conformance</strong><span>reports/conformance_matrix.md</span></li><li><strong>trust_report</strong><span>reports/security_trust_report.md</span></li><li><strong>python_compatibility</strong><span>reports/python_compatibility.md</span></li><li><strong>architecture_maintainability</strong><span>reports/architecture_maintainability.md</span></li><li><strong>permission_policy</strong><span>security/permission_policy.md</span></li><li><strong>runtime_permissions</strong><span>reports/runtime_permission_probes.md</span></li><li><strong>skill_atlas</strong><span>reports/skill_atlas.html</span></li><li><strong>compiled_targets</strong><span>reports/compiled_targets.md</span></li><li><strong>adoption_drift</strong><span>reports/adoption_drift_report.md</span></li><li><strong>review_waivers</strong><span>reports/review_waivers.md</span></li><li><strong>review_annotations</strong><span>reports/review_annotations.md</span></li><li><strong>world_class_evidence_plan</strong><span>reports/world_class_evidence_plan.md</span></li><li><strong>world_class_evidence_ledger</strong><span>reports/world_class_evidence_ledger.md</span></li><li><strong>world_class_evidence_intake</strong><span>reports/world_class_evidence_intake.md</span></li><li><strong>world_class_submission_review</strong><span>reports/world_class_submission_review.md</span></li><li><strong>world_class_operator_runbook</strong><span>reports/world_class_operator_runbook.md</span></li><li><strong>world_class_operator_runbook_html</strong><span>reports/world_class_operator_runbook.html</span></li><li><strong>world_class_claim_guard</strong><span>reports/world_class_claim_guard.md</span></li><li><strong>registry_audit</strong><span>reports/registry_audit.md</span></li><li><strong>package_verification</strong><span>reports/package_verification.md</span></li><li><strong>install_simulation</strong><span>reports/install_simulation.md</span></li><li><strong>upgrade_check</strong><span>reports/upgrade_check.md</span></li><li><strong>migration</strong><span>docs/migration-v2.md</span></li></ul>
|
|
</div>
|
|
</section>
|
|
|
|
<section id="trigger" class="twocol">
|
|
<div class="panel"><h2>触发实验</h2><p>13 trigger cases; 0 misroutes; 0 ambiguous</p></div>
|
|
<div class="panel"><h2>组合治理</h2><dl class='kv-grid'><div><dt>Skill 数</dt><dd>12</dd></div><div><dt>纳入审查</dt><dd>1</dd></div><div><dt>待处理冲突</dt><dd>0</dd></div><div><dt>待处理 owner</dt><dd>0</dd></div><div><dt>待处理过期</dt><dd>0</dd></div><div><dt>Telemetry Report Count</dt><dd>1</dd></div><div><dt>Actionable Drift Signal Count</dt><dd>0</dd></div><div><dt>Drift Signal Count</dt><dd>0</dd></div><div><dt>非行动项</dt><dd>24</dd></div></dl></div>
|
|
</section>
|
|
|
|
<section id="output" class="twocol">
|
|
<div class="panel"><h2>输出实验</h2><dl class='kv-grid'><div><dt>案例数</dt><dd>5</dd></div><div><dt>With Skill</dt><dd>100</dd></div><div><dt>Baseline</dt><dd>0</dd></div><div><dt>增益</dt><dd>100</dd></div><div><dt>Gate</dt><dd>是</dd></div></dl></div>
|
|
<div class="panel"><h2>执行证据</h2><dl class='kv-grid'><div><dt>运行数</dt><dd>10</dd></div><div><dt>命令执行</dt><dd>10</dd></div><div><dt>模型执行</dt><dd>0</dd></div><div><dt>记录样本</dt><dd>0</dd></div><div><dt>计时样本</dt><dd>10</dd></div><div><dt>估算 Token</dt><dd>10</dd></div></dl></div>
|
|
</section>
|
|
|
|
<section class="twocol">
|
|
<div class="panel"><h2>盲评包</h2><dl class='kv-grid'><div><dt>Pair Count</dt><dd>5</dd></div><div><dt>Answer Key Separate</dt><dd>是</dd></div><div><dt>With Skill Hidden Count</dt><dd>5</dd></div></dl></div>
|
|
<div class="panel"><h2>审定报告</h2><dl class='kv-grid'><div><dt>Pair Count</dt><dd>5</dd></div><div><dt>Judgment Count</dt><dd>0</dd></div><div><dt>待审</dt><dd>5</dd></div><div><dt>Agreement Count</dt><dd>0</dd></div><div><dt>Disagreement Count</dt><dd>0</dd></div><div><dt>Invalid Decision Count</dt><dd>0</dd></div><div><dt>答案揭示</dt><dd>0</dd></div><div><dt>答案隐藏</dt><dd>5</dd></div><div><dt>Reviewer Checklist Count</dt><dd>5</dd></div><div><dt>Reviewer Checklist Pending Count</dt><dd>5</dd></div></dl></div>
|
|
</section>
|
|
|
|
<section><h2>评审清单</h2><p class='muted'>先打开 reports/output_review_kit.md；每张卡片对应一个 blind A/B case，说明当前是否待判、答案是否仍隐藏、应填写的决策文件、有效字段和复跑命令。</p><div class='output-review-grid'><article class='output-review-card awaiting-decision'><div><span>awaiting-decision · 答案隐藏</span><h3>skill-package-contract</h3></div><p>Reviewer has not selected A or B yet; answer key remains hidden.</p><dl><dt>盲评包</dt><dd><code>reports/output_blind_review_pack.json</code></dd><dt>决策表</dt><dd><code>reports/output_review_decisions.json</code></dd><dt>提示词</dt><dd>Turn this repeated workflow into a reusable team skill package.</dd></dl><div class='output-review-steps'><div><h4>操作命令</h4><ul class='output-review-commands'><li><span>prepare_review_kit</span><code>python3 scripts/yao.py output-review-kit</code></li><li><span>write_template</span><code>python3 scripts/adjudicate_output_review.py --write-template</code></li><li><span>adjudicate</span><code>python3 scripts/yao.py output-review</code></li><li><span>refresh_review_studio</span><code>python3 scripts/yao.py review-studio .</code></li></ul></div><div><h4>字段要求</h4><ul class='output-review-fields'><li><strong>winner_variant</strong><span>A or B after reading only the blind review pack.</span></li><li><strong>confidence</strong><span>Optional number from 0 to 1.</span></li><li><strong>reason</strong><span>Short rationale; do not reveal baseline or with-skill labels before adjudication.</span></li></ul></div><div><h4>隐私边界</h4><ul><li>Do not paste raw private user data into the decision reason.</li><li>Do not open the answer key before reviewer choices are recorded.</li><li>Leave winner_variant blank when the reviewer is not ready to decide.</li></ul></div></div></article><article class='output-review-card awaiting-decision'><div><span>awaiting-decision · 答案隐藏</span><h3>output-eval-expectation</h3></div><p>Reviewer has not selected A or B yet; answer key remains hidden.</p><dl><dt>盲评包</dt><dd><code>reports/output_blind_review_pack.json</code></dd><dt>决策表</dt><dd><code>reports/output_review_decisions.json</code></dd><dt>提示词</dt><dd>Upgrade this production skill so we know whether the generated output is better than baseline.</dd></dl><div class='output-review-steps'><div><h4>操作命令</h4><ul class='output-review-commands'><li><span>prepare_review_kit</span><code>python3 scripts/yao.py output-review-kit</code></li><li><span>write_template</span><code>python3 scripts/adjudicate_output_review.py --write-template</code></li><li><span>adjudicate</span><code>python3 scripts/yao.py output-review</code></li><li><span>refresh_review_studio</span><code>python3 scripts/yao.py review-studio .</code></li></ul></div><div><h4>字段要求</h4><ul class='output-review-fields'><li><strong>winner_variant</strong><span>A or B after reading only the blind review pack.</span></li><li><strong>confidence</strong><span>Optional number from 0 to 1.</span></li><li><strong>reason</strong><span>Short rationale; do not reveal baseline or with-skill labels before adjudication.</span></li></ul></div><div><h4>隐私边界</h4><ul><li>Do not paste raw private user data into the decision reason.</li><li>Do not open the answer key before reviewer choices are recorded.</li><li>Leave winner_variant blank when the reviewer is not ready to decide.</li></ul></div></div></article><article class='output-review-card awaiting-decision'><div><span>awaiting-decision · 答案隐藏</span><h3>ir-before-packaging</h3></div><p>Reviewer has not selected A or B yet; answer key remains hidden.</p><dl><dt>盲评包</dt><dd><code>reports/output_blind_review_pack.json</code></dd><dt>决策表</dt><dd><code>reports/output_review_decisions.json</code></dd><dt>提示词</dt><dd>Package this skill for OpenAI, Claude, Agent Skills, and generic targets.</dd></dl><div class='output-review-steps'><div><h4>操作命令</h4><ul class='output-review-commands'><li><span>prepare_review_kit</span><code>python3 scripts/yao.py output-review-kit</code></li><li><span>write_template</span><code>python3 scripts/adjudicate_output_review.py --write-template</code></li><li><span>adjudicate</span><code>python3 scripts/yao.py output-review</code></li><li><span>refresh_review_studio</span><code>python3 scripts/yao.py review-studio .</code></li></ul></div><div><h4>字段要求</h4><ul class='output-review-fields'><li><strong>winner_variant</strong><span>A or B after reading only the blind review pack.</span></li><li><strong>confidence</strong><span>Optional number from 0 to 1.</span></li><li><strong>reason</strong><span>Short rationale; do not reveal baseline or with-skill labels before adjudication.</span></li></ul></div><div><h4>隐私边界</h4><ul><li>Do not paste raw private user data into the decision reason.</li><li>Do not open the answer key before reviewer choices are recorded.</li><li>Leave winner_variant blank when the reviewer is not ready to decide.</li></ul></div></div></article><article class='output-review-card awaiting-decision'><div><span>awaiting-decision · 答案隐藏</span><h3>near-neighbor-boundary</h3></div><p>Reviewer has not selected A or B yet; answer key remains hidden.</p><dl><dt>盲评包</dt><dd><code>reports/output_blind_review_pack.json</code></dd><dt>决策表</dt><dd><code>reports/output_review_decisions.json</code></dd><dt>提示词</dt><dd>I only need a one-off summary of these notes, not a reusable process.</dd></dl><div class='output-review-steps'><div><h4>操作命令</h4><ul class='output-review-commands'><li><span>prepare_review_kit</span><code>python3 scripts/yao.py output-review-kit</code></li><li><span>write_template</span><code>python3 scripts/adjudicate_output_review.py --write-template</code></li><li><span>adjudicate</span><code>python3 scripts/yao.py output-review</code></li><li><span>refresh_review_studio</span><code>python3 scripts/yao.py review-studio .</code></li></ul></div><div><h4>字段要求</h4><ul class='output-review-fields'><li><strong>winner_variant</strong><span>A or B after reading only the blind review pack.</span></li><li><strong>confidence</strong><span>Optional number from 0 to 1.</span></li><li><strong>reason</strong><span>Short rationale; do not reveal baseline or with-skill labels before adjudication.</span></li></ul></div><div><h4>隐私边界</h4><ul><li>Do not paste raw private user data into the decision reason.</li><li>Do not open the answer key before reviewer choices are recorded.</li><li>Leave winner_variant blank when the reviewer is not ready to decide.</li></ul></div></div></article><article class='output-review-card awaiting-decision'><div><span>awaiting-decision · 答案隐藏</span><h3>file-backed-governed-package</h3></div><p>Reviewer has not selected A or B yet; answer key remains hidden.</p><dl><dt>盲评包</dt><dd><code>reports/output_blind_review_pack.json</code></dd><dt>决策表</dt><dd><code>reports/output_review_decisions.json</code></dd><dt>提示词</dt><dd>Turn the attached release brief source into a governed skill package.</dd></dl><div class='output-review-steps'><div><h4>操作命令</h4><ul class='output-review-commands'><li><span>prepare_review_kit</span><code>python3 scripts/yao.py output-review-kit</code></li><li><span>write_template</span><code>python3 scripts/adjudicate_output_review.py --write-template</code></li><li><span>adjudicate</span><code>python3 scripts/yao.py output-review</code></li><li><span>refresh_review_studio</span><code>python3 scripts/yao.py review-studio .</code></li></ul></div><div><h4>字段要求</h4><ul class='output-review-fields'><li><strong>winner_variant</strong><span>A or B after reading only the blind review pack.</span></li><li><strong>confidence</strong><span>Optional number from 0 to 1.</span></li><li><strong>reason</strong><span>Short rationale; do not reveal baseline or with-skill labels before adjudication.</span></li></ul></div><div><h4>隐私边界</h4><ul><li>Do not paste raw private user data into the decision reason.</li><li>Do not open the answer key before reviewer choices are recorded.</li><li>Leave winner_variant blank when the reviewer is not ready to decide.</li></ul></div></div></article></div></section>
|
|
<section class="twocol">
|
|
<div class="panel"><h2>发布标准</h2><p>Governed 和 Library 至少需要 5 个 output eval cases，并覆盖 file-backed、near-neighbor、boundary case、execution evidence 和 blind A/B review pack。</p></div>
|
|
<div class="panel"><h2>人工结论</h2><p>没有 reviewer 决策时只显示 pending；只有真实决策文件会进入一致率和分歧统计。</p></div>
|
|
</section>
|
|
|
|
<section class="twocol">
|
|
<div class="panel"><h2>评审方式</h2><p>先打开 reports/output_review_kit.html 做盲评，填入 reports/output_review_decisions.json，再用 reports/output_review_adjudication.md 核对答案 key。</p></div>
|
|
<div class="panel"><h2>运行方式</h2><p>reports/output_execution_runs.md 会标明 recorded fixture、command run 或 model run；只有 provider runner 返回 model metadata 时才算 model-executed。</p></div>
|
|
</section>
|
|
|
|
<section id="runtime" class="twocol">
|
|
<div class="panel"><h2>运行矩阵</h2><dl class='kv-grid'><div><dt>目标数</dt><dd>5</dd></div><div><dt>通过数</dt><dd>5</dd></div><div><dt>Fail Count</dt><dd>0</dd></div></dl></div>
|
|
<div class="panel"><h2>目标编译</h2><dl class='kv-grid'><div><dt>目标数</dt><dd>5</dd></div><div><dt>通过数</dt><dd>5</dd></div><div><dt>Warn Count</dt><dd>0</dd></div><div><dt>Block Count</dt><dd>0</dd></div><div><dt>失败数</dt><dd>0</dd></div></dl></div>
|
|
</section>
|
|
|
|
<section class="twocol">
|
|
<div class="panel"><h2>上下文</h2><p>initial load 960/1000; deferred 416415/120000; top deferred scripts 367825; resource governance governed; quality density 135.4</p></div>
|
|
<div class="panel"><h2>编译证据</h2><p>Review reports/compiled_targets.md before packaging to inspect target adapter modes, generated files, preserved semantics, warnings, and unsupported features.</p></div>
|
|
</section>
|
|
|
|
<section id="trust" class="twocol">
|
|
<div class="panel"><h2>信任报告</h2><dl class='kv-grid'><div><dt>Secret</dt><dd>0</dd></div><div><dt>脚本数</dt><dd>106</dd></div><div><dt>网络脚本</dt><dd>3</dd></div><div><dt>Help 失败</dt><dd>0</dd></div><div><dt>包体哈希</dt><dd><code>28625ce8b3fcb3e214ef3b8ffc748d9cff551e4c5a1b78786c1a2b7d7005f679</code></dd></div></dl></div>
|
|
<div class="panel"><h2>安全边界</h2><p>高风险 secret、远程 inline execution、缺失依赖策略或无法解释的脚本接口应阻断 governed release。</p></div>
|
|
</section>
|
|
|
|
<section class="twocol">
|
|
<div class="panel"><h2>Python 兼容</h2><dl class='kv-grid'><div><dt>目标 Python</dt><dd>3.11</dd></div><div><dt>文件数</dt><dd>171</dd></div><div><dt>问题数</dt><dd>0</dd></div><div><dt>语法错误</dt><dd>0</dd></div><div><dt>F-string 3.11</dt><dd>0</dd></div></dl></div>
|
|
<div class="panel"><h2>解释器边界</h2><p>CI 和发布审查以 Python 3.11 兼容为底线；本地更高版本允许的新语法不能绕过兼容门禁。</p></div>
|
|
</section>
|
|
|
|
<section id="permissions" class="twocol">
|
|
<div class="panel"><h2>权限批准</h2><p>3/3 permissions approved; gaps 0; required file_write, network, subprocess</p></div>
|
|
<div class="panel"><h2>批准策略</h2><p>高权限能力需要 reviewer、scope、reason、expires_at 和 openai/claude/generic 目标端 enforcement 说明。</p></div>
|
|
</section>
|
|
|
|
<section id="permission-probes" class="twocol">
|
|
<div class="panel"><h2>权限探针</h2><p>4/4 targets probed; native 0; metadata fallback 4; installer 4; residual risks 4</p></div>
|
|
<div class="panel"><h2>运行边界</h2><dl class='kv-grid'><div><dt>目标数</dt><dd>4</dd></div><div><dt>通过数</dt><dd>4</dd></div><div><dt>原生执行</dt><dd>0</dd></div><div><dt>Metadata fallback</dt><dd>4</dd></div><div><dt>安装拦截</dt><dd>4</dd></div><div><dt>安装权限失败</dt><dd>0</dd></div><div><dt>残余风险</dt><dd>4</dd></div><div><dt>失败数</dt><dd>0</dd></div></dl></div>
|
|
</section>
|
|
|
|
<section id="atlas" class="twocol">
|
|
<div class="panel"><h2>组合治理</h2><p>12 skills, 1 actionable; 0 actionable route collisions; 0 actionable owner gaps; 0 actionable stale; 0 actionable drift; 24 scoped non-actionable issues</p></div>
|
|
<div class="panel"><h2>下一动作</h2><p>优先处理真实 portfolio 中的 duplicate names、stale skills、owner gaps，再用运营回路判断真实影响。</p></div>
|
|
</section>
|
|
|
|
<section id="telemetry" class="twocol">
|
|
<div class="panel"><h2>运营回路</h2><p>1 metadata events; adoption 100.0; missed 0; bad-output 0; risk low</p></div>
|
|
<div class="panel"><h2>漂移信号</h2><dl class='kv-grid'><div><dt>事件数</dt><dd>1</dd></div><div><dt>采用率</dt><dd>100</dd></div><div><dt>漏触发</dt><dd>0</dd></div><div><dt>Bad Output Count</dt><dd>0</dd></div><div><dt>风险带</dt><dd>low</dd></div></dl></div>
|
|
</section>
|
|
|
|
<section id="waivers">
|
|
<h2>人工批准</h2>
|
|
<p class="muted">warning 可以被有边界地接受，但必须写入 reviewer、理由、范围和到期时间；blocker 与 world-class 完成证据不能通过 waiver 变成通过。</p>
|
|
<div class="twocol waiver-summary">
|
|
<div class="panel"><h2>批准概况</h2><p>0 active waivers; 1 warning gates still need reviewer decision</p></div>
|
|
<div class="panel"><h2>批准台账</h2><dl class='kv-grid'><div><dt>Waiver Count</dt><dd>0</dd></div><div><dt>Active Count</dt><dd>0</dd></div><div><dt>Expired Count</dt><dd>0</dd></div><div><dt>Invalid Count</dt><dd>0</dd></div><div><dt>覆盖 Gate</dt><dd>0</dd></div></dl></div>
|
|
</div>
|
|
<h3 class="section-subtitle">批准候选</h3>
|
|
<div class='waiver-candidate-grid'><article class='waiver-card allowed'><div><span>可批准 · needs-reviewer-decision</span><h3>Output Lab</h3></div><p>review pending 5; model-executed 0; output failures 0</p><dl><dt>Gate</dt><dd><code>output-lab</code></dd><dt>证据</dt><dd><code>reports/output_review_adjudication.md</code></dd><dt>选项</dt><dd>accepted-risk, false-positive, temporary-exception</dd><dt>边界</dt><dd>Does not count as provider, human, or public world-class completion evidence.</dd></dl><div class='waiver-review'><h4>审查条件</h4><ul><li>Reviewer confirms this release does not claim provider-backed or human-adjudicated output superiority.</li><li>Reviewer names the release scope and expiry date.</li><li>Reviewer links output_review_adjudication or output_execution evidence.</li></ul></div><div class='waiver-command'><span>建议命令</span><code>python3 scripts/yao.py review-waivers . --add-waiver --gate-key output-lab --reviewer "<reviewer>" --reason "Output Lab has pending human/provider evidence; accepted only for this bounded review scope." --expires-at 2027-06-13 --evidence reports/output_review_adjudication.md</code></div></article><article class='waiver-card blocked'><div><span>不可批准 · cannot-waive</span><h3>World-Class Evidence</h3></div><p>4 pending evidence entries; 1 human pending; 3 external pending</p><dl><dt>Gate</dt><dd><code>world-class-evidence</code></dd><dt>证据</dt><dd><code>reports/world_class_evidence_ledger.md</code></dd><dt>选项</dt><dd>无</dd><dt>边界</dt><dd>Non-waivable completion boundary.</dd></dl><div class='waiver-review'><h4>审查条件</h4><ul><li>Do not use a waiver to claim public world-class readiness.</li><li>Either submit accepted ledger evidence or state that this release does not claim world-class completion.</li><li>Keep claim guard active until ledger summary.ready_to_claim_world_class is true.</li></ul></div><div class='waiver-command'><span>建议命令</span><code>python3 scripts/yao.py world-class-ledger . --submissions-dir evidence/world_class/submissions && python3 scripts/yao.py world-class-claim-guard .</code></div></article></div>
|
|
</section>
|
|
|
|
<section id="world-class">
|
|
<h2>世界证据</h2>
|
|
<p class="muted">这里列出每个 world-class 证据项的当前状态、完成定义、证据来源、隐私约束和下一步；计划、metadata fallback、待评审和本地命令不会被当成完成证据。</p>
|
|
<div class='world-evidence-grid'><article class='world-evidence-card pending'><div><span>待补证 · external</span><h3>Provider Holdout</h3></div><p>Collect at least one provider-backed output-eval holdout run with model, timing, and token metadata.</p><dl><dt>负责人</dt><dd>operator with provider credentials</dd><dt>当前状态</dt><dd>model-executed 0; token-observed 0</dd><dt>下一步</dt><dd>Run provider-backed holdout cases with real credentials and commit only aggregate evidence.</dd><dt>观测值</dt><dd>model_executed_count: 0; timing_observed_count: 10; token_observed_count: 0; accepted: False</dd><dt>提交态</dt><dd>status: missing; path: evidence/world_class/submissions/provider-holdout.json; attested_real_evidence: False; privacy_contract_satisfied: False</dd></dl><section class='world-runbook-panel'><h4>执行步骤</h4><ul><li>YAO_OUTPUT_EVAL_MODEL=gpt-4.1-mini OPENAI_API_KEY=<redacted> python3 scripts/yao.py output-exec --provider-runner openai --timeout-seconds 60</li><li>python3 scripts/yao.py skill-os2-audit . --generated-at <YYYY-MM-DD></li><li>Copy evidence/world_class/templates/provider-holdout.intake.json to evidence/world_class/submissions/provider-holdout.json and fill only real evidence fields.</li><li>python3 scripts/yao.py world-class-intake . --submissions-dir evidence/world_class/submissions</li></ul></section><div class='world-evidence-columns'><div><h4>完成定义</h4><ul><li>reports/output_execution_runs.json summary.model_executed_count > 0</li><li>reports/output_execution_runs.json summary.timing_observed_count > 0</li><li>reports/output_execution_runs.json summary.token_observed_count > 0</li><li>reports/skill_os2_audit.json item provider-holdout status becomes pass</li></ul></div><div><h4>证据来源</h4><ul><li>reports/output_execution_runs.json</li><li>reports/output_execution_runs.md</li><li>reports/skill_os2_audit.json</li><li>evidence/world_class/intake.schema.json</li><li>evidence/world_class/templates/provider-holdout.intake.json</li><li>reports/world_class_evidence_intake.json</li><li>reports/world_class_evidence_intake.md</li></ul></div><div><h4>隐私约束</h4><ul><li>Do not commit provider credentials or environment dumps.</li><li>The output execution report records output hashes and aggregate run metadata, not raw provider prompts.</li></ul></div></div><section class='world-source-panel'><h4>源证据检查</h4><ul class='world-source-checks'><li class='world-source-check blocked'><span>Provider model run</span><code>model_executed_count: 0 / >0</code><small>Run provider-backed output-exec with real credentials.</small></li><li class='world-source-check pass'><span>Timing observed</span><code>timing_observed_count: 10 / >0</code><small>Provider execution should record timing metadata.</small></li><li class='world-source-check blocked'><span>Token usage observed</span><code>token_observed_count: 0 / >0</code><small>Provider execution should return non-estimated token usage.</small></li></ul></section></article><article class='world-evidence-card pending'><div><span>待补证 · human</span><h3>Human Adjudication</h3></div><p>Record real blind A/B reviewer decisions before claiming human output review completion.</p><dl><dt>负责人</dt><dd>human reviewer</dd><dt>当前状态</dt><dd>0/5 decisions; pending 5</dd><dt>下一步</dt><dd>Record real A/B choices in the decision template, then regenerate adjudication.</dd><dt>观测值</dt><dd>pair_count: 5; judgment_count: 0; pending_count: 5; invalid_decision_count: 0; answer_revealed_count: 0; accepted: False</dd><dt>提交态</dt><dd>status: missing; path: evidence/world_class/submissions/human-adjudication.json; attested_real_evidence: False; privacy_contract_satisfied: False</dd></dl><section class='world-runbook-panel'><h4>执行步骤</h4><ul><li>python3 scripts/yao.py output-review-kit --write-template</li><li>Open reports/output_review_kit.md and choose A or B for each pair without opening the answer key.</li><li>python3 scripts/adjudicate_output_review.py --write-template</li><li>Edit reports/output_review_decisions.json with winner_variant values and reviewer metadata.</li><li>python3 scripts/yao.py output-review</li><li>python3 scripts/yao.py skill-os2-audit . --generated-at <YYYY-MM-DD></li><li>Copy evidence/world_class/templates/human-adjudication.intake.json to evidence/world_class/submissions/human-adjudication.json and fill only real evidence fields.</li><li>python3 scripts/yao.py world-class-intake . --submissions-dir evidence/world_class/submissions</li></ul></section><div class='world-evidence-columns'><div><h4>完成定义</h4><ul><li>reports/output_review_adjudication.json summary.pending_count == 0</li><li>reports/output_review_adjudication.json summary.judgment_count == summary.pair_count</li><li>reports/output_review_adjudication.json summary.invalid_decision_count == 0</li><li>reports/skill_os2_audit.json item human-adjudication status becomes pass</li></ul></div><div><h4>证据来源</h4><ul><li>reports/output_blind_review_pack.md</li><li>reports/output_review_kit.md</li><li>reports/output_review_decisions.json</li><li>reports/output_review_adjudication.json</li><li>reports/output_review_adjudication.md</li><li>evidence/world_class/intake.schema.json</li><li>evidence/world_class/templates/human-adjudication.intake.json</li><li>reports/world_class_evidence_intake.json</li><li>reports/world_class_evidence_intake.md</li></ul></div><div><h4>隐私约束</h4><ul><li>Reviewer decisions should not include raw user data or private customer detail.</li><li>Keep the answer key separate until after decisions are recorded.</li></ul></div></div><section class='world-source-panel'><h4>源证据检查</h4><ul class='world-source-checks'><li class='world-source-check pass'><span>Review pairs exist</span><code>pair_count: 5 / >0</code><small>Generate the blind A/B review pack.</small></li><li class='world-source-check blocked'><span>No pending decisions</span><code>pending_count: 5 / ==0</code><small>Record a reviewer choice for every pair.</small></li><li class='world-source-check blocked'><span>Judgments complete</span><code>judgment_count: 0 / ==pair_count</code><small>Every pair needs one valid human judgment.</small></li><li class='world-source-check pass'><span>No invalid decisions</span><code>invalid_decision_count: 0 / ==0</code><small>Fix malformed winner/confidence entries.</small></li></ul></section></article><article class='world-evidence-card pending'><div><span>待补证 · external</span><h3>Native Permission Enforcement</h3></div><p>Prove at least one real target client or external installer runtime guard enforces approved high-permission capabilities.</p><dl><dt>负责人</dt><dd>target client or installer integrator</dd><dt>当前状态</dt><dd>native-enforced targets 0; installer-enforced targets 4</dd><dt>下一步</dt><dd>Integrate a real target-client or external installer runtime guard before claiming native permission enforcement.</dd><dt>观测值</dt><dd>native_enforcement_count: 0; metadata_fallback_count: 4; installer_enforcement_pass_count: 4; installer_permission_failure_count: 0; installer_enforcement_ready: True; residual_risk_count: 4; failure_count: 0; accepted: False</dd><dt>提交态</dt><dd>status: missing; path: evidence/world_class/submissions/native-permission-enforcement.json; attested_real_evidence: False; privacy_contract_satisfied: False</dd></dl><section class='world-runbook-panel'><h4>执行步骤</h4><ul><li>Implement or connect a real target client or external installer runtime guard that blocks undeclared network, file_write, or subprocess capabilities.</li><li>Update the generated target adapter only when the guard is actually enforced by that target.</li><li>python3 scripts/yao.py package . --platform openai --platform claude --platform generic --platform vscode --output-dir dist --zip</li><li>python3 scripts/yao.py install-simulate . --package-dir dist --install-root dist/install-simulation</li><li>python3 scripts/yao.py runtime-permissions . --package-dir dist</li><li>python3 scripts/yao.py skill-os2-audit . --generated-at <YYYY-MM-DD></li><li>Copy evidence/world_class/templates/native-permission-enforcement.intake.json to evidence/world_class/submissions/native-permission-enforcement.json and fill only real evidence fields.</li><li>python3 scripts/yao.py world-class-intake . --submissions-dir evidence/world_class/submissions</li></ul></section><div class='world-evidence-columns'><div><h4>完成定义</h4><ul><li>reports/runtime_permission_probes.json summary.native_enforcement_count > 0</li><li>reports/runtime_permission_probes.json summary.failure_count == 0</li><li>reports/runtime_permission_probes.json summary.installer_enforcement_pass_count records local installer enforcement but does not replace native evidence</li><li>reports/skill_os2_audit.json item native-permission-enforcement status becomes pass</li></ul></div><div><h4>证据来源</h4><ul><li>dist/targets/*/adapter.json</li><li>reports/runtime_permission_probes.json</li><li>reports/runtime_permission_probes.md</li><li>reports/install_simulation.json</li><li>reports/install_simulation.md</li><li>security/permission_policy.json</li><li>evidence/world_class/intake.schema.json</li><li>evidence/world_class/templates/native-permission-enforcement.intake.json</li><li>reports/world_class_evidence_intake.json</li><li>reports/world_class_evidence_intake.md</li></ul></div><div><h4>隐私约束</h4><ul><li>Do not mark native_enforcement true for metadata-only fallbacks.</li><li>Keep residual risks visible for targets that still rely on operator enforcement.</li></ul></div></div><section class='world-source-panel'><h4>源证据检查</h4><ul class='world-source-checks'><li class='world-source-check blocked'><span>Native enforcement</span><code>native_enforcement_count: 0 / >0</code><small>Collect real target-client or external runtime guard proof.</small></li><li class='world-source-check pass'><span>Probe failures</span><code>failure_count: 0 / ==0</code><small>Runtime permission probes must stay clean.</small></li><li class='world-source-check pass'><span>Installer support</span><code>installer_enforcement_ready: True / true</code><small>Installer enforcement is supporting evidence, not native proof.</small></li></ul></section></article><article class='world-evidence-card pending'><div><span>待补证 · external</span><h3>Native Client Telemetry</h3></div><p>Import production metadata-only events from a real external client into the local drift loop.</p><dl><dt>负责人</dt><dd>Browser/Chrome/IDE/provider client integrator</dd><dt>当前状态</dt><dd>external source events 0; adoption samples 1</dd><dt>下一步</dt><dd>Install a real client against the native host and import production metadata-only events.</dd><dt>观测值</dt><dd>external_source_events: 0; adoption_sample_count: 1; raw_content_allowed: False; risk_band: low; accepted: False</dd><dt>提交态</dt><dd>status: missing; path: evidence/world_class/submissions/native-client-telemetry.json; attested_real_evidence: False; privacy_contract_satisfied: False</dd></dl><section class='world-runbook-panel'><h4>执行步骤</h4><ul><li>python3 scripts/telemetry_native_host.py . --write-launcher /tmp/yao-telemetry-host.sh --write-manifest /tmp/yao-telemetry-host.json --allowed-origin chrome-extension://<extension-id>/</li><li>Install the generated native messaging manifest for the real client and send at least one accepted skill_activation or skill_output event.</li><li>python3 scripts/yao.py telemetry-import . --input-jsonl .yao/telemetry_spool/external_events.jsonl</li><li>python3 scripts/yao.py skill-atlas --workspace-root .</li><li>python3 scripts/yao.py skill-os2-audit . --generated-at <YYYY-MM-DD></li><li>Copy evidence/world_class/templates/native-client-telemetry.intake.json to evidence/world_class/submissions/native-client-telemetry.json and fill only real evidence fields.</li><li>python3 scripts/yao.py world-class-intake . --submissions-dir evidence/world_class/submissions</li></ul></section><div class='world-evidence-columns'><div><h4>完成定义</h4><ul><li>reports/adoption_drift_report.json summary.source_types.external > 0</li><li>reports/adoption_drift_report.json summary.adoption_sample_count > 0</li><li>reports/skill_os2_audit.json item native-client-telemetry status becomes pass</li></ul></div><div><h4>证据来源</h4><ul><li>reports/adoption_drift_report.json</li><li>reports/adoption_drift_report.md</li><li>reports/telemetry_hook_recipes.json</li><li>scripts/telemetry_native_host.py</li><li>evidence/world_class/intake.schema.json</li><li>evidence/world_class/templates/native-client-telemetry.intake.json</li><li>reports/world_class_evidence_intake.json</li><li>reports/world_class_evidence_intake.md</li></ul></div><div><h4>隐私约束</h4><ul><li>Telemetry must remain metadata-only and local-first.</li><li>Do not package reports/telemetry_events.jsonl or any raw prompt, output, transcript, note, or message field.</li></ul></div></div><section class='world-source-panel'><h4>源证据检查</h4><ul class='world-source-checks'><li class='world-source-check blocked'><span>External events</span><code>external_source_events: 0 / >0</code><small>Import at least one metadata-only event from a real client.</small></li><li class='world-source-check pass'><span>Adoption sample</span><code>adoption_sample_count: 1 / >0</code><small>Telemetry must include adoption outcome evidence.</small></li><li class='world-source-check pass'><span>Raw content blocked</span><code>raw_content_allowed: False / false</code><small>Telemetry must stay metadata-only.</small></li></ul></section></article></div>
|
|
</section>
|
|
|
|
<section>
|
|
<h2>提交清单</h2>
|
|
<p class="muted">每张卡片给出模板、提交文件、准备命令、校验命令、收集要求、通过条件和隐私边界；只有真实 provider、真人、原生权限或真实客户端证据通过 intake 后，才进入 ledger review。</p>
|
|
<div class='world-intake-grid'><article class='world-intake-card awaiting-submission'><div><span>awaiting-submission · external</span><h3>Provider Holdout</h3></div><p>No real evidence submission has been provided yet.</p><dl><dt>负责人</dt><dd>operator with provider credentials</dd><dt>模板</dt><dd><code>evidence/world_class/templates/provider-holdout.intake.json</code></dd><dt>提交</dt><dd><code>evidence/world_class/submissions/provider-holdout.json</code></dd><dt>下一步</dt><dd>Run provider-backed holdout cases with real credentials and commit only aggregate evidence.</dd></dl><div class='world-intake-steps'><div><h4>操作命令</h4><ul class='world-intake-commands'><li><span>prepare_submission</span><code>python3 scripts/yao.py world-class-submission-kit . --evidence-key provider-holdout --output-dir evidence/world_class/submissions</code></li><li><span>validate_intake</span><code>python3 scripts/yao.py world-class-intake . --submissions-dir evidence/world_class/submissions</code></li><li><span>submission_review</span><code>python3 scripts/yao.py world-class-submission-review . --submissions-dir evidence/world_class/submissions</code></li><li><span>refresh_ledger</span><code>python3 scripts/yao.py world-class-ledger . --submissions-dir evidence/world_class/submissions</code></li><li><span>guard_claim</span><code>python3 scripts/yao.py world-class-claim-guard .</code></li></ul></div><div><h4>收集要求</h4><ul><li>provider-backed model run</li><li>observed timing</li><li>observed token metadata</li></ul></div><div><h4>执行步骤</h4><ul><li>YAO_OUTPUT_EVAL_MODEL=gpt-4.1-mini OPENAI_API_KEY=<redacted> python3 scripts/yao.py output-exec --provider-runner openai --timeout-seconds 60</li><li>python3 scripts/yao.py skill-os2-audit . --generated-at <YYYY-MM-DD></li><li>Copy evidence/world_class/templates/provider-holdout.intake.json to evidence/world_class/submissions/provider-holdout.json and fill only real evidence fields.</li><li>python3 scripts/yao.py world-class-intake . --submissions-dir evidence/world_class/submissions</li></ul></div><div><h4>通过条件</h4><ul><li>reports/output_execution_runs.json summary.model_executed_count > 0</li><li>reports/output_execution_runs.json summary.timing_observed_count > 0</li><li>reports/output_execution_runs.json summary.token_observed_count > 0</li><li>reports/skill_os2_audit.json item provider-holdout status becomes pass</li></ul></div><div><h4>隐私边界</h4><ul><li>Do not commit provider credentials or environment dumps.</li><li>The output execution report records output hashes and aggregate run metadata, not raw provider prompts.</li></ul></div></div></article><article class='world-intake-card awaiting-submission'><div><span>awaiting-submission · human</span><h3>Human Adjudication</h3></div><p>No real evidence submission has been provided yet.</p><dl><dt>负责人</dt><dd>human reviewer</dd><dt>模板</dt><dd><code>evidence/world_class/templates/human-adjudication.intake.json</code></dd><dt>提交</dt><dd><code>evidence/world_class/submissions/human-adjudication.json</code></dd><dt>下一步</dt><dd>Record real A/B choices in the decision template, then regenerate adjudication.</dd></dl><div class='world-intake-steps'><div><h4>操作命令</h4><ul class='world-intake-commands'><li><span>prepare_submission</span><code>python3 scripts/yao.py world-class-submission-kit . --evidence-key human-adjudication --output-dir evidence/world_class/submissions</code></li><li><span>validate_intake</span><code>python3 scripts/yao.py world-class-intake . --submissions-dir evidence/world_class/submissions</code></li><li><span>submission_review</span><code>python3 scripts/yao.py world-class-submission-review . --submissions-dir evidence/world_class/submissions</code></li><li><span>refresh_ledger</span><code>python3 scripts/yao.py world-class-ledger . --submissions-dir evidence/world_class/submissions</code></li><li><span>guard_claim</span><code>python3 scripts/yao.py world-class-claim-guard .</code></li></ul></div><div><h4>收集要求</h4><ul><li>real reviewer identity</li><li>blind A/B decisions</li><li>answer key unopened until decisions exist</li></ul></div><div><h4>执行步骤</h4><ul><li>python3 scripts/yao.py output-review-kit --write-template</li><li>Open reports/output_review_kit.md and choose A or B for each pair without opening the answer key.</li><li>python3 scripts/adjudicate_output_review.py --write-template</li><li>Edit reports/output_review_decisions.json with winner_variant values and reviewer metadata.</li><li>python3 scripts/yao.py output-review</li><li>python3 scripts/yao.py skill-os2-audit . --generated-at <YYYY-MM-DD></li><li>Copy evidence/world_class/templates/human-adjudication.intake.json to evidence/world_class/submissions/human-adjudication.json and fill only real evidence fields.</li><li>python3 scripts/yao.py world-class-intake . --submissions-dir evidence/world_class/submissions</li></ul></div><div><h4>通过条件</h4><ul><li>reports/output_review_adjudication.json summary.pending_count == 0</li><li>reports/output_review_adjudication.json summary.judgment_count == summary.pair_count</li><li>reports/output_review_adjudication.json summary.invalid_decision_count == 0</li><li>reports/skill_os2_audit.json item human-adjudication status becomes pass</li></ul></div><div><h4>隐私边界</h4><ul><li>Reviewer decisions should not include raw user data or private customer detail.</li><li>Keep the answer key separate until after decisions are recorded.</li></ul></div></div></article><article class='world-intake-card awaiting-submission'><div><span>awaiting-submission · external</span><h3>Native Permission Enforcement</h3></div><p>No real evidence submission has been provided yet.</p><dl><dt>负责人</dt><dd>target client or installer integrator</dd><dt>模板</dt><dd><code>evidence/world_class/templates/native-permission-enforcement.intake.json</code></dd><dt>提交</dt><dd><code>evidence/world_class/submissions/native-permission-enforcement.json</code></dd><dt>下一步</dt><dd>Integrate a real target-client or external installer runtime guard before claiming native permission enforcement.</dd></dl><div class='world-intake-steps'><div><h4>操作命令</h4><ul class='world-intake-commands'><li><span>prepare_submission</span><code>python3 scripts/yao.py world-class-submission-kit . --evidence-key native-permission-enforcement --output-dir evidence/world_class/submissions</code></li><li><span>validate_intake</span><code>python3 scripts/yao.py world-class-intake . --submissions-dir evidence/world_class/submissions</code></li><li><span>submission_review</span><code>python3 scripts/yao.py world-class-submission-review . --submissions-dir evidence/world_class/submissions</code></li><li><span>refresh_ledger</span><code>python3 scripts/yao.py world-class-ledger . --submissions-dir evidence/world_class/submissions</code></li><li><span>guard_claim</span><code>python3 scripts/yao.py world-class-claim-guard .</code></li></ul></div><div><h4>收集要求</h4><ul><li>real target client or external installer runtime guard</li><li>native enforcement flag or externally accepted guard proof</li><li>residual risk retained for fallback targets</li></ul></div><div><h4>执行步骤</h4><ul><li>Implement or connect a real target client or external installer runtime guard that blocks undeclared network, file_write, or subprocess capabilities.</li><li>Update the generated target adapter only when the guard is actually enforced by that target.</li><li>python3 scripts/yao.py package . --platform openai --platform claude --platform generic --platform vscode --output-dir dist --zip</li><li>python3 scripts/yao.py install-simulate . --package-dir dist --install-root dist/install-simulation</li><li>python3 scripts/yao.py runtime-permissions . --package-dir dist</li><li>python3 scripts/yao.py skill-os2-audit . --generated-at <YYYY-MM-DD></li><li>Copy evidence/world_class/templates/native-permission-enforcement.intake.json to evidence/world_class/submissions/native-permission-enforcement.json and fill only real evidence fields.</li><li>python3 scripts/yao.py world-class-intake . --submissions-dir evidence/world_class/submissions</li></ul></div><div><h4>通过条件</h4><ul><li>reports/runtime_permission_probes.json summary.native_enforcement_count > 0</li><li>reports/runtime_permission_probes.json summary.failure_count == 0</li><li>reports/runtime_permission_probes.json summary.installer_enforcement_pass_count records local installer enforcement but does not replace native evidence</li><li>reports/skill_os2_audit.json item native-permission-enforcement status becomes pass</li></ul></div><div><h4>隐私边界</h4><ul><li>Do not mark native_enforcement true for metadata-only fallbacks.</li><li>Keep residual risks visible for targets that still rely on operator enforcement.</li></ul></div></div></article><article class='world-intake-card awaiting-submission'><div><span>awaiting-submission · external</span><h3>Native Client Telemetry</h3></div><p>No real evidence submission has been provided yet.</p><dl><dt>负责人</dt><dd>Browser/Chrome/IDE/provider client integrator</dd><dt>模板</dt><dd><code>evidence/world_class/templates/native-client-telemetry.intake.json</code></dd><dt>提交</dt><dd><code>evidence/world_class/submissions/native-client-telemetry.json</code></dd><dt>下一步</dt><dd>Install a real client against the native host and import production metadata-only events.</dd></dl><div class='world-intake-steps'><div><h4>操作命令</h4><ul class='world-intake-commands'><li><span>prepare_submission</span><code>python3 scripts/yao.py world-class-submission-kit . --evidence-key native-client-telemetry --output-dir evidence/world_class/submissions</code></li><li><span>validate_intake</span><code>python3 scripts/yao.py world-class-intake . --submissions-dir evidence/world_class/submissions</code></li><li><span>submission_review</span><code>python3 scripts/yao.py world-class-submission-review . --submissions-dir evidence/world_class/submissions</code></li><li><span>refresh_ledger</span><code>python3 scripts/yao.py world-class-ledger . --submissions-dir evidence/world_class/submissions</code></li><li><span>guard_claim</span><code>python3 scripts/yao.py world-class-claim-guard .</code></li></ul></div><div><h4>收集要求</h4><ul><li>real external client source</li><li>metadata-only event</li><li>local-first import path</li></ul></div><div><h4>执行步骤</h4><ul><li>python3 scripts/telemetry_native_host.py . --write-launcher /tmp/yao-telemetry-host.sh --write-manifest /tmp/yao-telemetry-host.json --allowed-origin chrome-extension://<extension-id>/</li><li>Install the generated native messaging manifest for the real client and send at least one accepted skill_activation or skill_output event.</li><li>python3 scripts/yao.py telemetry-import . --input-jsonl .yao/telemetry_spool/external_events.jsonl</li><li>python3 scripts/yao.py skill-atlas --workspace-root .</li><li>python3 scripts/yao.py skill-os2-audit . --generated-at <YYYY-MM-DD></li><li>Copy evidence/world_class/templates/native-client-telemetry.intake.json to evidence/world_class/submissions/native-client-telemetry.json and fill only real evidence fields.</li><li>python3 scripts/yao.py world-class-intake . --submissions-dir evidence/world_class/submissions</li></ul></div><div><h4>通过条件</h4><ul><li>reports/adoption_drift_report.json summary.source_types.external > 0</li><li>reports/adoption_drift_report.json summary.adoption_sample_count > 0</li><li>reports/skill_os2_audit.json item native-client-telemetry status becomes pass</li></ul></div><div><h4>隐私边界</h4><ul><li>Telemetry must remain metadata-only and local-first.</li><li>Do not package reports/telemetry_events.jsonl or any raw prompt, output, transcript, note, or message field.</li></ul></div></div></article></div>
|
|
</section>
|
|
|
|
<section class="twocol">
|
|
<div class="panel"><h2>蓝图覆盖</h2><dl class='kv-grid'><div><dt>项目数</dt><dd>20</dd></div><div><dt>模块数</dt><dd>8</dd></div><div><dt>建议 PR</dt><dd>12</dd></div><div><dt>通过数</dt><dd>20</dd></div><div><dt>Warn Count</dt><dd>0</dd></div><div><dt>缺失数</dt><dd>0</dd></div><div><dt>Extension Track Count</dt><dd>2</dd></div><div><dt>Extension Covered Count</dt><dd>1</dd></div><div><dt>Extension Partial Count</dt><dd>1</dd></div><div><dt>Extension Planned Count</dt><dd>0</dd></div><div><dt>Adaptive Extension Ready</dt><dd>否</dd></div><div><dt>本地蓝图</dt><dd>是</dd></div><div><dt>世界级</dt><dd>否</dd></div><div><dt>待补证据</dt><dd>4</dd></div></dl></div>
|
|
<div class="panel"><h2>覆盖边界</h2><p>蓝图覆盖只证明 2.0 模块、建议 PR、脚本、报告和测试在本地闭环；public world-class 仍以 world-class evidence ledger 的真人和外部证据为准。</p></div>
|
|
</section>
|
|
|
|
<section class="twocol">
|
|
<div class="panel"><h2>公开声明</h2><dl class='kv-grid'><div><dt>本地复现</dt><dd>是</dd></div><div><dt>发布锁</dt><dd>否</dd></div><div><dt>可公开声明</dt><dd>否</dd></div><div><dt>声明阻断</dt><dd>5</dd></div><div><dt>Provider 证据</dt><dd>否</dd></div><div><dt>人审完成</dt><dd>否</dd></div><div><dt>世界级就绪</dt><dd>否</dd></div></dl></div>
|
|
<div class="panel"><h2>声明阻断</h2><ul class='issues'><li><strong>阻断</strong><span>release lock is not clean or commit is unavailable</span></li><li><strong>阻断</strong><span>provider-backed model holdout evidence is incomplete</span></li><li><strong>阻断</strong><span>human blind-review adjudication is incomplete</span></li><li><strong>阻断</strong><span>world-class evidence is not accepted yet (4 open gaps, 4 ledger pending)</span></li><li><strong>阻断</strong><span>world-class source checks are not all accepted (6/13 pass, 7 blocked)</span></li></ul></div>
|
|
</section>
|
|
|
|
<section class="twocol">
|
|
<div class="panel"><h2>世界证据</h2><p>4 pending world-class evidence entries; 1 human pending; 3 external pending; source checks 7/13 pass; 6 blocked; overclaim guard true</p></div>
|
|
<div class="panel"><h2>证据台账</h2><dl class='kv-grid'><div><dt>Ledger Entry Count</dt><dd>4</dd></div><div><dt>Accepted Count</dt><dd>0</dd></div><div><dt>待审</dt><dd>4</dd></div><div><dt>Human Pending Count</dt><dd>1</dd></div><div><dt>External Pending Count</dt><dd>3</dd></div><div><dt>Overclaim Guard Active</dt><dd>是</dd></div><div><dt>Ready To Claim World Class</dt><dd>否</dd></div></dl></div>
|
|
</section>
|
|
|
|
<section class="twocol">
|
|
<div class="panel"><h2>证据入口</h2><dl class='kv-grid'><div><dt>Schema</dt><dd>是</dd></div><div><dt>模板数</dt><dd>4</dd></div><div><dt>模板通过</dt><dd>4</dd></div><div><dt>提交数</dt><dd>0</dd></div><div><dt>有效提交</dt><dd>0</dd></div><div><dt>无效提交</dt><dd>0</dd></div><div><dt>可收证据</dt><dd>是</dd></div><div><dt>可审台账</dt><dd>否</dd></div><div><dt>Ready To Claim World Class</dt><dd>否</dd></div></dl></div>
|
|
<div class="panel"><h2>入口边界</h2><p>intake 只校验证据包格式、来源、隐私和反过度声明；只有 ledger 看到真实 provider、真人、原生权限或真实客户端结果后，world-class 才能进入完成审计。</p></div>
|
|
</section>
|
|
|
|
<section class="twocol">
|
|
<div class="panel"><h2>声明守卫</h2><dl class='kv-grid'><div><dt>台账可声明</dt><dd>否</dd></div><div><dt>台账待补</dt><dd>4</dd></div><div><dt>声明面</dt><dd>76</dd></div><div><dt>违规数</dt><dd>0</dd></div><div><dt>Overclaim Guard Active</dt><dd>是</dd></div></dl></div>
|
|
<div class="panel"><h2>声明边界</h2><p>claim guard 扫描 README、docs 和 reports 中的完成态表述；ledger 未 ready 时，任何英文完成断言、true 状态声明或中文完成态都会阻断发布审查。</p></div>
|
|
</section>
|
|
|
|
<section id="registry" class="twocol">
|
|
<div class="panel"><h2>注册审计</h2><p>yao-meta-skill 1.1.0; 6/6 compatibility entries pass; install pass with 4 adapters; installer permissions 12 enforced / 0 failures</p></div>
|
|
<div class="panel"><h2>包体元数据</h2><dl class='kv-grid'><div><dt>名称</dt><dd>yao-meta-skill</dd></div><div><dt>版本</dt><dd>1.1.0</dd></div><div><dt>Maturity</dt><dd>governed</dd></div><div><dt>Owner</dt><dd>Yao Team</dd></div><div><dt>License</dt><dd>MIT</dd></div><div><dt>信任级别</dt><dd>local</dd></div><div><dt>目标平台</dt><dd>openai, claude, generic, agent-skills-compatible, vscode</dd></div><div><dt>兼容通过</dt><dd>6/6</dd></div><div><dt>归档哈希</dt><dd><code>6852cf91a74d232c32d732b7c159c971827abf23af50153987193b084ad3b5cc</code></dd></div></dl></div>
|
|
</section>
|
|
|
|
<section id="release" class="twocol">
|
|
<div class="panel"><h2>发布路线</h2><p>0 promote; 3 keep current; 0 blocked; upgrade minor declared / minor recommended</p></div>
|
|
<div class="panel"><h2>包体验证</h2><dl class='kv-grid'><div><dt>目标数</dt><dd>4</dd></div><div><dt>Adapter</dt><dd>4</dd></div><div><dt>归档存在</dt><dd>是</dd></div><div><dt>Zip 条目</dt><dd>586</dd></div><div><dt>失败数</dt><dd>0</dd></div><div><dt>警告数</dt><dd>0</dd></div><div><dt>归档哈希</dt><dd><code>6852cf91a74d232c32d732b7c159c971827abf23af50153987193b084ad3b5cc</code></dd></div></dl></div>
|
|
</section>
|
|
|
|
<section class="twocol">
|
|
<div class="panel"><h2>安装模拟</h2><dl class='kv-grid'><div><dt>Archive Extracted</dt><dd>是</dd></div><div><dt>Entrypoint Loaded</dt><dd>是</dd></div><div><dt>Manifest Loaded</dt><dd>是</dd></div><div><dt>Interface Loaded</dt><dd>是</dd></div><div><dt>Adapter</dt><dd>4</dd></div><div><dt>安装权限</dt><dd>12</dd></div><div><dt>安装权限失败</dt><dd>0</dd></div><div><dt>权限目标</dt><dd>4</dd></div><div><dt>权限能力</dt><dd>3</dd></div><div><dt>失败数</dt><dd>0</dd></div><div><dt>警告数</dt><dd>0</dd></div></dl></div>
|
|
<div class="panel"><h2>权限覆盖</h2><p>安装模拟会读取解压后的 target adapter，并按 declared_capabilities 核对 security/permission_policy.json 中的有效批准、证据、过期时间和目标端 enforcement 说明。</p></div>
|
|
</section>
|
|
</main>
|
|
</body>
|
|
</html>
|