4 targets in platform-neutral contract
Review Studio 2.0
@@ -133,65 +214,121 @@
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.
审查结论
- review
- Score 87/100
+ ready
+ Score 100/100
核心指标
-5 cases; 1 file-backed
target conformance pass rate
50 scripts scanned; secrets found
12 scanned skills; route collisions
4 targets in platform-neutral contract
target contracts compiled from Skill IR
5 cases; 1 file-backed
command 10; model 0; recorded 0
review pairs hide baseline vs with-skill labels
adjudication decisions; pending 5
target conformance pass rate
0 native enforcement targets
62 scripts scanned; secrets found
12 scanned skills; route collisions
1 metadata events; 0 missed triggers
0 gates covered; human risk decisions
0 open blocker annotations
4 targets; MIT license
483 zip entries; package verification
3 adapters readable; local install simulation
declared minor; 0 breaking changes
审查闸门
-关注
意图画布
intent confidence 30/100; Pause before deep authoring and close the highest-leverage gaps first.
通过
触发实验
13 trigger cases; 0 misroutes; 0 ambiguous
通过
输出实验
5/5 cases; with-skill 100.0; baseline 0.0; file-backed 1; near-neighbor 1
通过
上下文
initial load 910/1000; quality density 142.9
通过
运行矩阵
5 / 5 targets pass
关注
信任报告
0 secrets; 50 scripts; 2 network-capable scripts
关注
组合治理
12 skills; 5 route collisions; 9 owner gaps; 10 stale
通过
发布路线
0 promote; 3 keep current; 0 blocked
通过
意图画布
intent confidence 100/100; Intent is clear enough to package the first routeable version.
通过
触发实验
13 trigger cases; 0 misroutes; 0 ambiguous
通过
输出实验
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
通过
上下文
initial load 987/1000; quality density 131.7
通过
运行矩阵
5 / 5 targets pass
通过
信任报告
0 secrets; 62 scripts; 2 network-capable scripts; 0 help smoke failures
通过
权限批准
3/3 permissions approved; gaps 0; required file_write, network, subprocess
通过
权限探针
3/3 targets probed; native 0; metadata fallback 3; residual risks 3
通过
组合治理
12 skills, 1 actionable; 0 actionable route collisions; 0 actionable owner gaps; 0 actionable stale; 24 scoped non-actionable issues
通过
运营回路
1 metadata events; adoption 100.0; missed 0; bad-output 0; risk low
通过
人工批准
0 active waivers cover current warnings
通过
注册审计
yao-meta-skill 1.1.0; 6/6 compatibility entries pass; install pass with 3 adapters
通过
发布路线
0 promote; 3 keep current; 0 blocked; upgrade minor declared / minor recommended
阻断事项
无。
关注事项
- 意图画布intent confidence 30/100; Pause before deep authoring and close the highest-leverage gaps first.
- 信任报告0 secrets; 50 scripts; 2 network-capable scripts
- 组合治理12 skills; 5 route collisions; 9 owner gaps; 10 stale
关注事项
无。
修复动作
+当前没有 blocker 或 warning。保持现有证据链即可。
审查批注
+当前批注：0 条批注；开放 0；阻断 0
+当前没有 reviewer 批注。
意图画布
-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.
+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.
证据路径
-- skill_overviewreports/skill-overview.html
- review_viewerreports/review-viewer.html
- output_evalreports/output_quality_scorecard.md
- runtime_conformancereports/conformance_matrix.md
- trust_reportreports/security_trust_report.md
- skill_atlasreports/skill_atlas.html
- migrationdocs/migration-v2.md
- skill_overviewreports/skill-overview.html
- review_viewerreports/review-viewer.html
- output_evalreports/output_quality_scorecard.md
- output_executionreports/output_execution_runs.md
- output_blind_reviewreports/output_blind_review_pack.md
- output_review_adjudicationreports/output_review_adjudication.md
- runtime_conformancereports/conformance_matrix.md
- trust_reportreports/security_trust_report.md
- permission_policysecurity/permission_policy.md
- runtime_permissionsreports/runtime_permission_probes.md
- skill_atlasreports/skill_atlas.html
- compiled_targetsreports/compiled_targets.md
- adoption_driftreports/adoption_drift_report.md
- review_waiversreports/review_waivers.md
- review_annotationsreports/review_annotations.md
- registry_auditreports/registry_audit.md
- package_verificationreports/package_verification.md
- install_simulationreports/install_simulation.md
- upgrade_checkreports/upgrade_check.md
- migrationdocs/migration-v2.md
触发实验
13 trigger cases; 0 misroutes; 0 ambiguous
组合治理
{'skill_count': 12, 'route_collision_count': 5, 'owner_gap_count': 9, 'stale_count': 10, 'shared_resource_count': 0, 'no_route_opportunity_count': 0}
组合治理
{'skill_count': 12, 'actionable_skill_count': 1, 'route_collision_count': 5, 'actionable_route_collision_count': 0, 'owner_gap_count': 9, 'actionable_owner_gap_count': 0, 'stale_count': 10, 'actionable_stale_count': 0, 'shared_resource_count': 0, 'no_route_opportunity_count': 0, 'non_actionable_issue_count': 24}
输出实验
{'case_count': 5, 'file_backed_case_count': 1, 'near_neighbor_case_count': 1, 'boundary_case_count': 1, 'baseline_pass_rate': 0.0, 'with_skill_pass_rate': 100.0, 'delta': 100.0, 'regression_count': 0, 'gate_pass': True, 'failure_taxonomy': []}
发布标准
Governed 和 Library 至少需要 5 个 output eval cases，并覆盖 file-backed、near-neighbor 和 boundary case。
输出实验
{'case_count': 5, 'file_backed_case_count': 1, 'near_neighbor_case_count': 1, 'boundary_case_count': 1, 'baseline_pass_rate': 0.0, 'with_skill_pass_rate': 100.0, 'delta': 100.0, 'regression_count': 0, 'gate_pass': True, 'failure_taxonomy': [], 'blind_pair_count': 5}
执行证据
{'case_count': 5, 'variant_run_count': 10, 'command_executed_count': 10, 'model_executed_count': 0, 'recorded_fixture_count': 0, 'timing_observed_count': 10, 'token_observed_count': 0, 'token_estimated_count': 10, 'baseline_pass_rate': 0.0, 'with_skill_pass_rate': 100.0, 'delta': 100.0, 'regression_count': 0, 'failure_count': 0, 'gate_pass': True}
盲评包
{'pair_count': 5, 'answer_key_separate': True, 'with_skill_hidden_count': 5}
审定报告
{'pair_count': 5, 'judgment_count': 0, 'pending_count': 5, 'agreement_count': 0, 'disagreement_count': 0, 'invalid_decision_count': 0, 'agreement_rate': None, 'needs_review': True, 'failure_count': 0}
发布标准
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。
人工结论
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。
评审方式
先看 reports/output_blind_review_pack.md 做盲评，填入 reports/output_review_decisions.json，再用 reports/output_review_adjudication.md 核对答案 key。
运行方式
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。
运行矩阵
{'target_count': 5, 'pass_count': 5, 'fail_count': 0}
上下文
initial load 910/1000; quality density 142.9
目标编译
{'target_count': 4, 'pass_count': 4, 'warn_count': 0, 'block_count': 0, 'failure_count': 0, 'warning_count': 0}
上下文
initial load 987/1000; quality density 131.7
编译证据
Review reports/compiled_targets.md before packaging to inspect target adapter modes, generated files, preserved semantics, warnings, and unsupported features.
信任报告
{'scanned_files': 127, 'script_count': 50, 'secret_findings': 0, 'dependency_files': ['requirements-ci.txt'], 'network_script_count': 2, 'interactive_script_count': 0, 'package_sha256': '18618ecee3683f7488ae4a65a9800c428ee15c6539bbcd367910c57d84ac77ad'}
信任报告
{'scanned_files': 146, 'script_count': 62, 'internal_module_count': 3, 'secret_findings': 0, 'dependency_files': ['requirements-ci.txt'], 'network_script_count': 2, 'network_policy_covered_count': 2, 'network_policy_missing_count': 0, 'file_write_script_count': 47, 'permission_required_count': 3, 'permission_approved_count': 3, 'permission_missing_count': 0, 'permission_invalid_count': 0, 'permission_expired_count': 0, 'help_smoke_checked_count': 59, 'help_smoke_failed_count': 0, 'interactive_script_count': 0, 'package_hash_scope': 'source-contract-without-generated-reports', 'package_hash_file_count': 146, 'package_sha256': '9f1d53dbaeb1c90c80d7ffbbc1d10473e50b8032acda68332fbae551bb246abe'}
安全边界
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。
权限批准
3/3 permissions approved; gaps 0; required file_write, network, subprocess
批准策略
高权限能力需要 reviewer、scope、reason、expires_at 和 openai/claude/generic 目标端 enforcement 说明。
权限探针
3/3 targets probed; native 0; metadata fallback 3; residual risks 3
运行边界
{'target_count': 3, 'pass_count': 3, 'fail_count': 0, 'native_enforcement_count': 0, 'metadata_fallback_count': 3, 'residual_risk_count': 3, 'required_capability_count': 3, 'failure_count': 0}
组合治理
12 skills; 5 route collisions; 9 owner gaps; 10 stale
下一动作
优先处理真实 portfolio 中的 duplicate names、stale skills、owner gaps，再做 registry 或 telemetry。
组合治理
12 skills, 1 actionable; 0 actionable route collisions; 0 actionable owner gaps; 0 actionable stale; 24 scoped non-actionable issues
下一动作
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。
运营回路
1 metadata events; adoption 100.0; missed 0; bad-output 0; risk low
漂移信号
{'event_count': 1, 'activation_count': 1, 'accepted_count': 1, 'edited_count': 0, 'rejected_count': 0, 'missed_count': 0, 'failed_count': 0, 'adoption_rate': 100.0, 'missed_trigger_count': 0, 'wrong_trigger_count': 0, 'bad_output_count': 0, 'script_error_count': 0, 'missing_resource_count': 0, 'review_overdue_count': 0, 'risk_band': 'low', 'event_types': {'skill_activation': 1}, 'failure_types': {}}
人工批准
0 active waivers cover current warnings
批准台账
{'waiver_count': 0, 'active_count': 0, 'expired_count': 0, 'invalid_count': 0, 'covered_gate_count': 0, 'covered_gate_keys': []}
注册审计
yao-meta-skill 1.1.0; 6/6 compatibility entries pass; install pass with 3 adapters
包体元数据
{'schema_version': '2.0', 'name': 'yao-meta-skill', 'version': '1.1.0', 'description': '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.', 'targets': ['openai', 'claude', 'generic', 'agent-skills-compatible'], 'maturity': 'governed', 'owner': 'Yao Team', 'review_cadence': 'quarterly', 'trust_level': 'local', 'license': 'MIT', 'checksums': {'package_sha256': '9f1d53dbaeb1c90c80d7ffbbc1d10473e50b8032acda68332fbae551bb246abe', 'archive_sha256': '4d1227dc0996a97aa4987761d46081391fe15b36362e19765e1e7aa8e5d41c06'}, 'compatibility': {'openai': 'pass', 'claude': 'pass', 'agent-skills': 'pass', 'vscode': 'pass', 'generic': 'pass', 'agent-skills-compatible': 'pass'}, 'source': {'skill_dir': '.', 'skill_ir': 'skill-ir/examples/yao-meta-skill.json', 'ir_schema_version': '2.0.0', 'canonical_metadata': 'agents/interface.yaml'}, 'artifacts': {'overview_html': 'reports/skill-overview.html', 'review_studio_html': 'reports/review-studio.html', 'trust_report': 'reports/security_trust_report.md', 'conformance_matrix': 'reports/conformance_matrix.md', 'compiled_targets': 'reports/compiled_targets.md', 'package_verification': 'reports/package_verification.md', 'install_simulation': 'reports/install_simulation.md', 'adoption_drift': 'reports/adoption_drift_report.md', 'review_waivers': 'reports/review_waivers.md', 'review_annotations': 'reports/review_annotations.md', 'package_metadata': 'registry/packages/yao-meta-skill.json'}, 'distribution': {'archive_verified': True, 'archive_sha256': '4d1227dc0996a97aa4987761d46081391fe15b36362e19765e1e7aa8e5d41c06', 'package_verification': 'reports/package_verification.json', 'install_simulated': True, 'install_simulation': 'reports/install_simulation.json'}, 'generated_at': '2026-06-13'}
发布路线
0 promote; 3 keep current; 0 blocked
世界级缺口
下一阶段应继续推进 IR-first compiler、registry audit、telemetry drift loop，以及更严格的 governed trust gates。
发布路线
0 promote; 3 keep current; 0 blocked; upgrade minor declared / minor recommended
包体验证
{'target_count': 3, 'adapter_count': 3, 'archive_present': True, 'archive_sha256': '4d1227dc0996a97aa4987761d46081391fe15b36362e19765e1e7aa8e5d41c06', 'archive_entry_count': 483, 'failure_count': 0, 'warning_count': 0}