sanook-cli 0.5.1 → 0.5.2

package/second-brain/Sessions/2026-06-18-final-gate-template-final.md

@@ -0,0 +1,192 @@
+---
+tags: [final-gate, verification, second-brain]
+note_type: final-gate
+created: 2026-06-18
+updated: 2026-06-18
+parent: "[[Sessions/_Index]]"
+---
+
+# 2026-06-18 - Final Gate Template - Final Gate
+
+> Evidence-backed closeout for adding [[Templates/final]] and wiring it into the Sanook second-brain framework.
+
+## 0. Final Gate Rule
+
+- [x] This final gate was created before the final owner-facing answer.
+- [x] Every PASS/PARTIAL/FAIL/BLOCKED claim below has evidence or an explicit reason.
+- [x] No unchecked item is silently treated as done.
+- [x] If a row has no evidence, it cannot be `PASS`.
+
+## 1. Objective / DoD Lock
+
+Original request:
+
+```text
+งั้น ช่วยทำทั้งแปดข้อให้ฉันหน่อยอย่างละเอียด
+```
+
+Expected output:
+
+- [x] Create a detailed `final.md` template that implements the eight proposed logic areas.
+- [x] Wire the template into existing second-brain indexes/standards so agents can discover it.
+- [x] Add verification that the template is bundled and keeps the eight sections.
+
+Definition of Done:
+
+| DoD item | Status | Evidence | Notes |
+|---|---|---|---|
+| `Templates/final.md` exists and covers all eight logic areas. | PASS | `second-brain/Templates/final.md`; `npm test -- src/brain.test.ts src/brain-review.test.ts` | Test asserts all eight headings. |
+| Template is linked from canonical indexes/hot path. | PASS | `Templates/_Index.md`, `Shared/AI-Context-Index.md`, `verification-standard.md`, operating sequence | Discoverable from template, hot context, and technical standard. |
+| Existing scaffold copies final template. | PASS | `src/brain.test.ts`; targeted test passed | `scaffoldBrain` copies bundled markdown seeds. |
+| Work was verified without touching unrelated gateway changes. | PASS | `git status --short`; final response note | Gateway files were pre-existing unrelated changes and were not edited. |
+
+## 2. Evidence-Backed Checklist
+
+| Gate | Status | Evidence | Notes |
+|---|---|---|---|
+| Relevant source/context was read before editing. | PASS | Read template/index/verification files and `src/brain.test.ts` | Used current repo state, not memory only. |
+| Existing canonical note/file was searched before creating a new one. | PASS | `find second-brain/Templates...`; `rg final/checklist/verification` | No existing `Templates/final.md`. |
+| Changes were made in the correct folder/module boundary. | PASS | `Templates/final.md`; Tech-Standards/Runbook/AI index updates | No root folder added. |
+| No unrelated user/worktree changes were reverted or overwritten. | PASS | `git status --short` still shows gateway files separate | Gateway files untouched. |
+| New/changed durable notes have `parent` and `up::`. | PASS | `brain.test.ts` bundled markdown hygiene test passed | Includes final/session notes. |
+| Index/backlink/discoverability was updated. | PASS | `Templates/_Index.md`, `Sessions/_Index.md`, `Shared/Tech-Standards/_Index.md` | Added links. |
+| Verification was run at the right scope. | PASS | Targeted tests, typecheck, full tests, diff check | Listed below. |
+| Residual risk is stated clearly. | PASS | Residual Risk section | No CLI command implemented yet. |
+
+## 3. Status Matrix
+
+| Work item / phase | Status | Evidence | Owner-visible outcome |
+|---|---|---|---|
+| Frame objective and DoD | PASS | This final gate section 1 | The eight items became concrete deliverables. |
+| Retrieve required context | PASS | Read local template/standard/runbook files | Work fits existing vault taxonomy. |
+| Implement / write / edit | PASS | Files changed list below | `final.md` and wiring added. |
+| Verify behavior or artifact | PASS | `npm test`, typecheck, targeted tests | Scaffold/template hygiene preserved. |
+| Update memory/index/session if needed | PASS | Session note + Sessions index + current-state updates | Trace preserved. |
+| Prepare final response | PASS | Final-answer draft below | Response will match evidence. |
+
+## 4. Evidence Matrix
+
+Commands:
+
+| Command | Status | Important output | Scope proven |
+|---|---|---|---|
+| `npm test -- src/brain.test.ts src/brain-review.test.ts` | PASS | 2 files, 26 tests passed | Template/scaffold/review targeted behavior. |
+| `npm run typecheck` | PASS | `tsc --noEmit` passed | TypeScript compile surface. |
+| `git diff --check` | PASS | No output | No whitespace conflict markers. |
+| `npm test` | PASS | 100 files, 865 tests passed | Full current test suite. |
+
+Files inspected:
+
+| File/path | Why it matters | Evidence |
+|---|---|---|
+| `second-brain/Shared/Tech-Standards/verification-standard.md` | Existing DoD and verification contract | Updated with Final Gate section. |
+| `second-brain/Checklists/preflight-postflight-template.md` | Existing pre/postflight checklist pattern | Final template complements, not replaces it. |
+| `second-brain/Templates/session.md` | Existing session closeout structure | Final gate links to session/handoff closeout. |
+| `second-brain/Runbooks/ai-second-brain-operating-sequence.md` | Agent phase model | Added final gate note under Eval. |
+| `src/brain.test.ts` | Scaffold/template regression guard | Added final gate assertions. |
+
+Files changed:
+
+| File/path | Change summary | Evidence |
+|---|---|---|
+| `second-brain/Templates/final.md` | New canonical final gate template with eight detailed sections. | File exists and targeted tests passed. |
+| `second-brain/Templates/_Index.md` | Links final template with other templates. | Index updated. |
+| `second-brain/Shared/Tech-Standards/verification-standard.md` | Adds final gate contract. | Standard updated. |
+| `second-brain/Shared/Tech-Standards/_Index.md` | Links verification standard. | Index updated. |
+| `second-brain/Shared/AI-Context-Index.md` | Adds hot-path pointer to final template. | Index updated. |
+| `second-brain/Runbooks/ai-second-brain-operating-sequence.md` | Adds final gate in Eval phase. | Runbook updated. |
+| `second-brain/Sessions/2026-06-18-final-gate-template.md` | Session log for this work. | File exists. |
+| `second-brain/Sessions/2026-06-18-final-gate-template-final.md` | This final gate. | File exists. |
+| `src/brain.test.ts` | Regression tests for final template. | Targeted tests passed. |
+
+## 5. Residual Risk
+
+| Risk | Impact | Mitigation / next check |
+|---|---|---|
+| No `sanook brain final` command exists yet. | Agents/users must instantiate `Templates/final.md` manually. | Dogfood manually; automate later if repeated use proves the shape. |
+| Template is detailed and may be too heavy for trivial tasks. | Could create checklist theater if overused. | Use only for non-trivial, multi-file, framework, release, or auditable tasks. |
+| Full tests passed with unrelated gateway worktree changes present. | Test result covers current worktree, not a clean branch. | Final answer should mention gateway files were unrelated and not touched. |
+
+Checks not run:
+
+| Check | Reason not run | Consequence |
+|---|---|---|
+| `npm run build` | This change is markdown/test-only; typecheck/full tests passed. | Build artifact not separately regenerated for this note-only change. |
+| `sanook brain review/eval` | Not required for a template-only framework note, and current configured vault may be user-local. | Could run later after indexing the actual vault. |
+
+## 6. Change Summary Audit
+
+What changed:
+
+- Added a detailed final gate template.
+- Wired it into template index, AI context index, verification standard, and operating sequence.
+- Added tests to keep the final template bundled and structurally complete.
+- Logged the work in Sessions and current-state.
+
+What did not change:
+
+- No CLI command was added.
+- No root folder was added.
+- Existing unrelated gateway files were not edited.
+
+Behavior before:
+
+- The vault had preflight/postflight and session templates, but no canonical final evidence matrix.
+
+Behavior after:
+
+- Non-trivial work can instantiate [[Templates/final]] before the final answer and close with evidence instead of unchecked claims.
+
+## 7. Final Answer Draft
+
+```text
+ทำทั้ง 8 ข้อเป็น `Templates/final.md` แล้ว และผูกเข้ากับ verification standard, AI hot path, operating sequence, indexes, session log, current-state, และ scaffold tests.
+
+Verified:
+- npm test -- src/brain.test.ts src/brain-review.test.ts: PASS
+- npm run typecheck: PASS
+- git diff --check: PASS
+- npm test: PASS (100 files, 865 tests)
+
+Residual risk:
+- ยังไม่ได้ทำ CLI automation เช่น `sanook brain final`; ตอนนี้เป็น reusable template/manual gate ก่อน
+```
+
+## 8. Second-Brain Routing / Memory Closeout
+
+Session / handoff:
+
+- [x] Added [[Sessions/2026-06-18-final-gate-template]].
+- [x] Added this final gate note.
+- [x] Updated [[Sessions/_Index]].
+
+Durable memory:
+
+- [x] Updated [[Shared/Operating-State/current-state]] with final gate status.
+- [x] No user preference or personal invariant was discovered.
+
+Indexes / discoverability:
+
+- [x] Updated [[Templates/_Index]].
+- [x] Updated [[Shared/Tech-Standards/_Index]].
+- [x] Updated [[Shared/AI-Context-Index]].
+
+Quality loop:
+
+- [x] Targeted and full tests passed.
+- [x] Final gate template is covered by `src/brain.test.ts`.
+
+## Final Verdict
+
+| Verdict | Choose one | Evidence |
+|---|---|---|
+| Ready to close | YES | Full tests, typecheck, diff check, targeted scaffold tests passed. |
+| Close with caveats | YES | No CLI automation yet; template/manual gate only. |
+| Needs more work | NO | Requested eight detailed logic areas are implemented in template. |
+| Blocked | NO | No blocker. |
+
+One-line final state:
+
+> Ready to close: `Templates/final.md` implements all eight requested closeout logic areas with evidence requirements and is wired into the vault framework.
+
+up:: [[Sessions/_Index]]

package/second-brain/Sessions/2026-06-18-final-gate-template.md

@@ -0,0 +1,71 @@
+---
+tags: [session, session-log, final-gate, verification, second-brain]
+note_type: session-log
+created: 2026-06-18
+updated: 2026-06-18
+parent: "[[Sessions/_Index]]"
+ai_surface: history
+---
+
+# 2026-06-18 - Final Gate Template
+
+> Session log for turning the owner's `final.md` idea into a reusable evidence-backed closeout template.
+
+## Summary
+
+Created [[Templates/final]] as the final phase gate for non-trivial work. It turns the owner's eight requested logic areas into a detailed checklist with required evidence, status vocabulary, residual risk, change audit, final answer draft, and second-brain closeout routing.
+
+## What Was Tried
+
+- Read existing verification and closeout notes:
+  - [[Shared/Tech-Standards/verification-standard]]
+  - [[Checklists/preflight-postflight-template]]
+  - [[Templates/session]]
+  - [[Runbooks/ai-second-brain-operating-sequence]]
+  - [[Shared/Rules/rules-formatting]]
+- Added `Templates/final.md`.
+- Linked it from [[Templates/_Index]].
+- Added final-gate rules to [[Shared/Tech-Standards/verification-standard]].
+- Added hot-path pointers in [[Shared/AI-Context-Index]] and [[Runbooks/ai-second-brain-operating-sequence]].
+- Added scaffold tests so the final template stays bundled and keeps the eight logic sections.
+
+## Errors
+
+- None blocking.
+
+## Solutions
+
+- Final gate uses `PASS`, `PARTIAL`, `FAIL`, `N/A`, and `BLOCKED`.
+- Every row requires evidence; no-evidence rows cannot be `PASS`.
+- Final answer draft is generated from evidence rather than memory or intent.
+- Closeout includes session/handoff, durable memory routing, indexes, and Sanook brain doctor/review/eval gates.
+
+## Key Decisions
+
+- `Templates/final.md` is the canonical reusable template.
+- Actual final gates should be instantiated as session/project artifacts when needed; the template itself stays in `Templates/`.
+- `final.md` should remain read-only guidance for now, not a new CLI command, until repeated use proves an automation shape.
+
+## Files Changed
+
+- `second-brain/Templates/final.md`
+- `second-brain/Templates/_Index.md`
+- `second-brain/Shared/Tech-Standards/verification-standard.md`
+- `second-brain/Shared/Tech-Standards/_Index.md`
+- `second-brain/Shared/AI-Context-Index.md`
+- `second-brain/Runbooks/ai-second-brain-operating-sequence.md`
+- `src/brain.test.ts`
+
+## Verification
+
+- `npm test -- src/brain.test.ts src/brain-review.test.ts`: PASS
+- `npm run typecheck`: PASS
+- `git diff --check`: PASS
+- `npm test`: PASS (100 files, 865 tests)
+
+## Next Steps
+
+- [ ] Dogfood [[Templates/final]] on the next non-trivial implementation task.
+- [ ] If it proves useful repeatedly, consider `sanook brain final` or `sanook brain new final`.
+
+up:: [[Sessions/_Index]]

package/second-brain/Sessions/2026-06-18-framework-dogfood-permission-and-memory.md

@@ -0,0 +1,58 @@
+---
+tags: [session, session-log, second-brain, dogfood, permission]
+note_type: session-log
+created: 2026-06-18
+updated: 2026-06-18
+parent: "[[Sessions/_Index]]"
+ai_surface: history
+---
+
+# 2026-06-18 — Framework Dogfood: Permission + Memory
+
+> Session evidence for using the new AI framework/context packs on a real permission-gate task plus durable memory updates.
+
+## Summary
+
+ใช้ framework ใหม่กับงานจริง 3 task-family ใน session เดียว: coding-release, second-brain maintenance, และ research/framework benchmark.
+
+## What Was Tried
+
+- **Coding & Release Work:** อ่าน diff ของ `src/tools/permission.ts` และ `src/tools/tools.test.ts`, แล้วรัน targeted test
+- **Second-Brain Maintenance:** อัปเดต [[Shared/User-Memory/user-preferences]] และ [[Shared/User-Memory/response-examples]] จาก feedback "ทำทั้งหมดได้เลย"
+- **Research To Framework / Benchmark:** รัน `Shared/Scripts/ai-second-brain-method-eval.mjs` และใช้ [[Evals/second-brain-benchmarks]] เป็น self-check
+
+## Errors
+
+- ไม่มี test failure ใน targeted permission/tools test
+- ไม่มี whitespace diff issue ตอนตรวจ `git diff --check`
+
+## Solutions
+
+- Permission/tools targeted test ผ่าน: `npm test -- src/tools/tools.test.ts`
+- เพิ่ม preference: เมื่อ scope ชัดและปลอดภัย ให้ทำครบตามรายการที่แนะนำได้เลย
+- เพิ่ม response example: "Act On All Clear Recommendations"
+- บันทึก benchmark/dogfood result ลง [[Evals/quality-ledger]]
+
+## Key Decisions
+
+- นับ session นี้เป็น dogfood ของ context packs 3 แบบ:
+  - [[Shared/Context-Packs/coding-release]]
+  - [[Shared/Context-Packs/second-brain-maintenance]]
+  - [[Shared/Context-Packs/research-to-framework]]
+- ยังไม่แก้ logic ใน `src/tools/permission.ts` เพราะ targeted tests ผ่านและ diff สอดคล้องกับเจตนาของ permission gate
+
+## Files Changed
+
+- `second-brain/Shared/User-Memory/user-preferences.md`
+- `second-brain/Shared/User-Memory/response-examples.md`
+- `second-brain/Evals/quality-ledger.md`
+- `second-brain/Shared/Operating-State/current-state.md`
+- `second-brain/Sessions/_Index.md`
+- `second-brain/Sessions/2026-06-18-framework-dogfood-permission-and-memory.md`
+
+## Next Steps
+
+- รัน full test/typecheck ถ้าจะ finalize code change ใน permission gate
+- ใช้ benchmark อีกครั้งหลัง permission changes ถูก merge หรือปรับเพิ่ม
+
+up:: [[Sessions/_Index]]

package/second-brain/Sessions/2026-06-18-hermes-second-brain-expansion-research.md

@@ -0,0 +1,52 @@
+---
+tags: [session, session-log, second-brain, hermes, research]
+note_type: session-log
+created: 2026-06-18
+updated: 2026-06-18
+parent: "[[Sessions/_Index]]"
+ai_surface: history
+---
+
+# 2026-06-18 — Hermes Second-Brain Expansion Research
+
+> Session log for the research pass on what to add to this second-brain for Hermes CLI support. Links to the sourced research note and leaves implementation as a next step.
+
+## Summary
+
+Researched Hermes Agent docs, context engineering guidance, agent memory papers, LangGraph memory concepts, and PARA. Main conclusion: avoid many new root folders; prioritize Hermes-specific adapter/context, memory sync policy, atomic benchmarks, golden fixtures, and scheduled vault-health reviews.
+
+## What Was Tried
+
+- Read current vault map, AI context index, evals index, context packs index, user memory index, and operating sequence.
+- Checked current worktree status and avoided unrelated CLI/search changes.
+- Searched current Hermes docs for context files, persistent memory, skills, and curator behavior.
+- Compared with external agent-memory/context sources.
+
+## Errors
+
+- Some docs pages were rendered HTML-heavy; raw GitHub markdown for Hermes docs was easier to inspect.
+
+## Solutions
+
+- Added [[Research/2026-06-18-hermes-cli-second-brain-expansion-research]] as the durable research note with source URLs and prioritized recommendations.
+
+## Key Decisions
+
+- No taxonomy folders were created yet.
+- Recommended next build is `HERMES.md` plus a small Hermes-specific shared policy area before wider taxonomy changes.
+
+## Files Changed
+
+- `second-brain/Research/2026-06-18-hermes-cli-second-brain-expansion-research.md`
+- `second-brain/Sessions/2026-06-18-hermes-second-brain-expansion-research.md`
+- `second-brain/Research/_Index.md`
+- `second-brain/Sessions/_Index.md`
+- `second-brain/Shared/Operating-State/current-state.md`
+
+## Next Steps
+
+- [ ] Decide whether to implement the recommended Hermes bundle.
+- [ ] If adding folders, update `Vault Structure Map.md` and `src/brain.ts` together.
+- [ ] If only adding files under existing folders, keep taxonomy unchanged.
+
+up:: [[Sessions/_Index]]

package/second-brain/Sessions/2026-06-18-mcp-ecosystem-and-sanook-ux-scan.md

@@ -0,0 +1,81 @@
+---
+tags: [session, session-log, mcp, sanook-cli]
+note_type: session-log
+created: 2026-06-18
+updated: 2026-06-18
+parent: "[[Sessions/_Index]]"
+ai_surface: history
+---
+
+# 2026-06-18 - MCP Ecosystem and Sanook UX Scan
+
+> Session log for reviewing Sanook's MCP support and current MCP registry integration opportunities.
+
+## Summary
+
+- Inspected `src/mcp.ts`, `src/mcp-server.ts`, `src/bin.ts`, `README.md`, and existing MCP tests/docs.
+- Queried the official MCP registry API and OpenAPI metadata.
+- Identified that Sanook already supports stdio + remote Streamable HTTP and can act as an MCP server.
+- Main UX gap: no registry-backed search/info/install/test wizard yet.
+- Added [[Research/2026-06-18-sanook-mcp-ecosystem-and-ux-roadmap]] and [[Shared/Tech-Standards/mcp-integration-roadmap]].
+- Implemented the P0 MCP UX inside Sanook CLI: registry search/info/install, presets, list/test/doctor probing, docs, and tests.
+- Verified local stdio MCP probing works; official GitLab remote installs from registry but returns `401 Unauthorized` without an auth header.
+
+## What Was Tried
+
+- Checked current MCP code paths with `rg`.
+- Read MCP client/server implementation.
+- Queried registry endpoints:
+  - `https://registry.modelcontextprotocol.io/v0/version`
+  - `https://registry.modelcontextprotocol.io/v0/health`
+  - `https://registry.modelcontextprotocol.io/v0/servers?...`
+  - `https://registry.modelcontextprotocol.io/openapi.json`
+- Sampled registry search terms: GitHub, GitLab, Postgres, SQLite, filesystem, Slack, Discord, Notion, Linear, Jira, Sentry, Playwright, browser, Gmail, Drive, Obsidian, Context7, fetch, Brave, Docker, Kubernetes.
+
+## Errors
+
+- Initial unquoted `curl` query with `?limit=8` was expanded by zsh; quoted URL fixed it.
+- Mintlify HTML docs are noisy for terminal extraction; the registry OpenAPI and API responses were more useful as machine-readable evidence.
+
+## Solutions
+
+- Used quoted registry URLs.
+- Used the registry OpenAPI for endpoint/field shape.
+- Captured a Sanook-specific roadmap instead of implementing a large installer in the research turn.
+- Added `src/mcp-registry.ts` for official registry parsing and install-plan generation.
+- Added `sanook mcp search`, `info`, `install`, `preset`, `test`, `doctor`, and `list --tools`.
+- Kept unit tests network-free; live registry smoke was run separately against built `dist/bin.js`.
+- Smoke result: `mcp search/info/install/list` worked against the official registry; a fake stdio MCP server passed `mcp test`; GitLab remote probe failed with `401 Unauthorized` until a token/header is supplied.
+
+## Key Decisions
+
+- Next best implementation should be `sanook mcp search/info/install/test`, not another hardcoded list.
+- Use registry metadata for env/header prompts and secret handling.
+- Keep project-local MCP behind `sanook trust`.
+- Prefer read-only / remote Streamable HTTP defaults where available.
+- Do not make the regular test suite depend on the live registry.
+- Preserve manual `--header` support because some registry entries do not yet declare required auth metadata.
+
+## Files Changed
+
+- `second-brain/Research/2026-06-18-sanook-mcp-ecosystem-and-ux-roadmap.md`
+- `second-brain/Shared/Tech-Standards/mcp-integration-roadmap.md`
+- `second-brain/Shared/Tech-Standards/_Index.md`
+- `second-brain/Research/_Index.md`
+- `second-brain/Sessions/_Index.md`
+- `second-brain/Sessions/2026-06-18-mcp-ecosystem-and-sanook-ux-scan.md`
+- `src/mcp-registry.ts`
+- `src/mcp-registry.test.ts`
+- `src/mcp.ts`
+- `src/mcp.test.ts`
+- `src/bin.ts`
+- `README.md`
+- `CHANGELOG.md`
+
+## Next Steps
+
+- Improve auth hints for hosted MCPs that return `401` but do not declare header requirements in registry metadata.
+- Add risk labels / read-write warnings before installing high-impact servers.
+- Consider registry result caching and richer install previews.
+
+up:: [[Sessions/_Index]]

package/second-brain/Sessions/2026-06-18-sanook-brain-cli-p0-implementation.md

@@ -0,0 +1,86 @@
+---
+tags: [session, session-log, sanook-cli, second-brain, implementation]
+note_type: session-log
+created: 2026-06-18
+updated: 2026-06-18
+parent: "[[Sessions/_Index]]"
+ai_surface: history
+---
+
+# 2026-06-18 - Sanook Brain CLI P0 Implementation
+
+> Implementation session for making the second-brain workflow first-class inside Sanook CLI.
+
+## Summary
+
+Implemented Sanook-native P0 commands for second-brain operations:
+
+- `sanook brain doctor`
+- `sanook brain context [--task "..."]`
+- `sanook brain eval`
+- `sanook brain review`
+
+## What Was Tried
+
+- Verified existing `brain doctor` implementation and tests.
+- Refactored `buildBrainContext()` so the exact prompt context is assembled from inspectable source parts.
+- Added a `brain context` module and CLI runner to show context source status, char counts, stale/missing warnings, and optional task retrieval.
+- Added a `brain eval` module and CLI runner that scores `Evals/second-brain-benchmarks.md` through static vault checks, context checks, index freshness, and retrieval probes.
+- Added a `brain review` module and CLI runner for read-only curator checks over Memory-Inbox, Context-Packs, Sessions/index coverage, eval freshness, and markdown routing hygiene.
+
+## Errors
+
+- The existing Memory-Inbox helper intentionally swallowed missing-file errors for runtime safety; `brain context` needed to distinguish `missing` from `empty`. Fixed by separating candidate parsing from file reading.
+- Review of a fresh scaffold surfaced that generated `Shared/Context-Packs/_Index.md` did not link bundled packs. Fixed scaffold generation so new vaults are discoverable immediately.
+- Review hygiene initially flagged root adapter files for missing `parent`; adjusted review to match the existing root-file exception rule.
+
+## Solutions
+
+- `src/memory.ts` now exports `buildBrainContextParts()` and `renderBrainContext()`.
+- `src/brain-context.ts` provides parser, inspector, and formatter for `sanook brain context`.
+- `src/brain-eval.ts` provides static and retrieval benchmark checks for `sanook brain eval`.
+- `src/brain-review.ts` provides read-only curator checks and formatter for `sanook brain review`.
+- `src/brain.ts` now renders seed context pack links into generated `Shared/Context-Packs/_Index.md`.
+- CLI help and REPL help now mention the new commands.
+
+## Files Changed
+
+- `CHANGELOG.md`
+- `src/bin.ts`
+- `src/commands.ts`
+- `src/memory.ts`
+- `src/brain.ts`
+- `src/brain-context.ts`
+- `src/brain-context.test.ts`
+- `src/brain-eval.ts`
+- `src/brain-eval.test.ts`
+- `src/brain-review.ts`
+- `src/brain-review.test.ts`
+- `second-brain/Projects/sanook-cli/second-brain-feature-roadmap.md`
+- `second-brain/Shared/Operating-State/current-state.md`
+
+## Verification
+
+- `npm test -- src/brain-eval.test.ts src/brain-context.test.ts src/memory.test.ts src/brain-doctor.test.ts src/brain.test.ts`
+- `npm run typecheck`
+- `npm test`
+- `npm run build`
+- `git diff --check`
+- Smoke with temp `HOME`: `brain init`, `index`, `brain doctor`, `brain context --task ... --no-content`, `brain eval --no-retrieval`
+- Smoke with temp `HOME`: `brain eval` full retrieval probes passed 15.0/15
+- `npm test -- src/brain-review.test.ts src/brain.test.ts src/memory.test.ts`
+- `npm run typecheck`
+- `npm test -- src/brain-review.test.ts src/brain-eval.test.ts src/brain-context.test.ts src/brain-doctor.test.ts src/brain.test.ts src/memory.test.ts`
+- `npm test`
+- `npm run build`
+- `git diff --check`
+- Smoke with temp `HOME`: `brain init`, `index`, `brain review` passed 0 warnings / 0 failures.
+- `node dist/bin.js --help | rg "brain (review|context|eval|doctor)"`
+
+## Next Steps
+
+- [ ] Consider `sanook brain pack list|show` for `Shared/Context-Packs/`.
+- [ ] Consider `sanook brain new <type>`.
+- [ ] Consider `sanook brain repair`.
+
+up:: [[Sessions/_Index]]