chore: import upstream snapshot with attribution

cli-tool/components/loops/evaluation/devils-advocate-loop.md

@@ -0,0 +1,42 @@
+---
+name: devils-advocate-loop
+description: Adversarially challenges a design or decision until every serious objection is resolved or explicitly accepted.
+category: evaluation
+interval: 15m
+stop-condition: No unresolved serious objections remain — each is fixed or consciously accepted with rationale.
+components: [agent:expert-advisors/architect-review, agent:expert-advisors/critical-thinking, command:git-workflow/pr-review]
+tags: [evaluation, design-review, decision-making, loop]
+---
+
+# Devil's-Advocate Loop
+
+> **Loop Engineering** — put a dedicated critic in the loop. It keeps attacking the design until the proposal survives scrutiny or is knowingly accepted with its trade-offs.
+
+## 🎯 Goal
+Stress-test a design, RFC, or architectural decision by generating the strongest possible objections, then resolving or explicitly accepting each one — no objection silently ignored.
+
+## ⏱️ Schedule
+Suggested interval: `15m` (or run until the design is settled).
+
+## ▶️ Run it
+```
+/loop 15m "Act as a devil's advocate against the current design. Raise the strongest objections (scaling, security, cost, edge cases, maintainability). For each, either fix the design or record an explicit accepted trade-off with rationale. Continue until no serious objection is unresolved."
+```
+
+## 🔁 Iteration steps
+1. **Perceive** — read the current design/decision.
+2. **Reason** — the `critical-thinking` agent generates the strongest counter-arguments.
+3. **Plan** — for each objection, choose fix vs accept-with-rationale.
+4. **Act** — update the design or the decision log; `architect-review` validates the revision.
+5. **Observe** — re-run with `/pr-review`; stop only when nothing serious is unresolved.
+
+## 🛑 Stopping condition
+Every serious objection is either resolved or recorded as an explicit, justified trade-off.
+
+## 🧩 Referenced components
+- `agent:expert-advisors/architect-review` — validates each revision.
+- `agent:expert-advisors/critical-thinking` — generates adversarial objections.
+- `command:git-workflow/pr-review` — structured review of the design changes.
+
+## 💡 Example
+A proposed single-region database is challenged on availability; the loop either adds a replication plan or records "single-region accepted for MVP, revisit at scale" with reasoning.

cli-tool/components/loops/evaluation/human-approval-loop.md

@@ -0,0 +1,45 @@
+---
+name: human-approval-loop
+description: Runs the task, then pauses and sends you approve / revise / skip before anything ships — human review as a first-class queue where the stop condition is your approval.
+category: evaluation
+interval: on-demand
+stop-condition: You approve the result (it ships) or skip/redirect it — nothing ships without explicit human approval.
+components: [hook:quality-gates/plan-gate, hook:automation/telegram-notifications, command:git-workflow/create-pr]
+tags: [human-in-the-loop, approval, safety, loop]
+---
+
+# Human-in-the-Loop Approval Loop
+
+> **Loop Engineering** — same loop shape as the autonomous ones, but the stop condition is *your approval* instead of a passing test. The agent does the work, then waits at a gate: approve, revise, or skip.
+
+## 🎯 Goal
+Run the task, then **pause** and route the result to you for a decision before anything ships — treating human review as its own queue with clear actions.
+
+## ⏱️ Schedule
+Trigger: `on-demand` (each item pauses for your call before shipping).
+
+## ▶️ Run it
+```
+/loop run the next task, then pause and send me approve / revise / skip before anything ships. On approve, continue and ship. On revise, take my note and redo. On skip, move to the next item. Never ship without my approval.
+```
+
+## 🔁 Iteration steps
+1. **Act** — complete the task up to the point of shipping.
+2. **Gate** — the `plan-gate` hook halts before any irreversible/ship step.
+3. **Ask** — send approve / revise / skip via `telegram-notifications`.
+4. **Branch** — approve → ship (e.g. `/create-pr`); revise → apply note and redo; skip → next item.
+5. **Observe** — record the decision; nothing ships unattended.
+
+## 🛑 Stopping condition
+You approve (ship) or skip/redirect — the loop never ships without an explicit human decision.
+
+## 💰 Budget & guardrails
+The gate is the safety: irreversible actions wait for a human. Add a reminder/deadline so pending items don't pile up silently, and a budget for the work done before each gate.
+
+## 🧩 Referenced components
+- `hook:quality-gates/plan-gate` — halts before irreversible/ship steps.
+- `hook:automation/telegram-notifications` — sends the approve/revise/skip prompt.
+- `command:git-workflow/create-pr` — the ship action on approval.
+
+## 💡 Example
+The agent drafts a refund-policy change, pauses, and pings you on Telegram. You hit "revise" with a note; it redoes the draft and pings again — and only ships after you approve.

cli-tool/components/loops/evaluation/quality-streak-loop.md

@@ -0,0 +1,42 @@
+---
+name: quality-streak-loop
+description: Runs realistic scenarios and fixes failures until the suite passes N times in a row with no regressions.
+category: evaluation
+interval: 20m
+stop-condition: The scenario suite passes N consecutive times (e.g. 5) with zero failures.
+components: [agent:performance-testing/test-automator, hook:testing/test-runner, command:testing/test-quality-analyzer]
+tags: [evaluation, reliability, testing, loop]
+---
+
+# Quality Streak Loop
+
+> **Loop Engineering** — don't trust a single green run. Require a *streak* of clean passes to prove stability before stopping.
+
+## 🎯 Goal
+Exercise the product with realistic scenarios, fix every failure, and keep going until the suite passes N consecutive times — catching flakiness and intermittent bugs.
+
+## ⏱️ Schedule
+Suggested interval: `20m`.
+
+## ▶️ Run it
+```
+/loop 20m "Run the realistic scenario suite. Fix any failure you find, including flaky ones. Reset the streak counter on any failure. Continue until the suite passes 5 times in a row."
+```
+
+## 🔁 Iteration steps
+1. **Perceive** — run the scenario suite via the `test-runner` hook.
+2. **Reason** — analyze failures (real bug vs flake) with `/test-quality-analyzer`.
+3. **Plan** — fix the failure and harden the test.
+4. **Act** — implement the fix with `test-automator`; reset the streak on any failure.
+5. **Observe** — repeat; only stop after N consecutive clean runs.
+
+## 🛑 Stopping condition
+N consecutive passes (default 5) with zero failures or flakes.
+
+## 🧩 Referenced components
+- `agent:performance-testing/test-automator` — fixes and hardens tests.
+- `hook:testing/test-runner` — auto-runs the suite.
+- `command:testing/test-quality-analyzer` — distinguishes real bugs from flakes.
+
+## 💡 Example
+The suite passes 4 times then a timing-dependent test flakes; the loop stabilizes it with a deterministic clock and restarts the streak until it hits 5 clean.