feat(bstack): doctor + frontmatter + primitives.md + templates for P13#6
Merged
Conversation
…3 (Dream Cycle Discipline) Companion to broomva/workspace#45 (P13 governance). Closes the doc gap where references/primitives.md and assets/templates/* hadn't been updated since P11 — now covers P12 (Persistent Loop) AND the new P13 (Dream Cycle). scripts/doctor.sh: - EXPECTED_COUNT: 12 → 13 - Count header regex: "Twelve/12" → "Thirteen/13" - Primitive row check loop: 1..12 → 1..13 - P_NAMES[]: append "P13: Dream Cycle Discipline" - REFLEXIVE_PRIMS[]: P6 P7 P10 P11 P12 → + P13 SKILL.md: - Frontmatter: P1–P12 → P1–P13 - "Twelve irreducible" → "Thirteen" - New P13 row in primitives quick-reference (closes shadow-dream corruption mode) - Reflexive primitives list now includes P13 references/primitives.md (closes a doc gap that was open since P11): - Added full P12 section (Persistent Loop Discipline) with reflexive trigger rule — was missing entirely - Added full P13 section (Dream Cycle Discipline) with 5-phase canonical shape table, implementation matrix, reflexive trigger rule, rule-of-three deferral note for the morpheus crate - Cohesion narrative updated: P11/P12/P13 are structural siblings at different scales — table comparing discipline / surface / evidence / scale for each - Closing autonomous-loop diagram updated with "IF long-horizon → P12" and "P13 dream cycle for any consolidation" - Table of contents updated (P12, P13 entries) assets/templates/CLAUDE.md.template: - "Eleven irreducible" → "Thirteen" - New P12 + P13 rows in the primitives table assets/templates/AGENTS.md.template: - New P12 section with reflexive trigger rule (was missing) - New P13 section with reflexive trigger rule - Closing diagram updated Smoke test (per P11 trigger #3): Against post-PR-α worktree: 46/49 passes 3 gaps are P6/P7/P12 skill-script paths (skills live at workspace root, not inside worktrees — expected, same pattern as prior doctor extensions) After PR α merges + npx skills update propagates to a real installed workspace, doctor reports full primitive contract compliance. This is the four-PR series for P13: - workspace#45 (P13 governance) — open - this PR (bstack doctor + frontmatter + templates + primitives ref) - broomva/bookkeeping#1 (P6 with replay — already merged) was the first concrete dream instance that motivates P13's binding now Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
|
Warning Rate limit exceeded
To continue reviewing without waiting, purchase usage credits in the billing tab. ⌛ How to resolve this issue?After the wait time has elapsed, a review can be triggered using the We recommend that you space out your commits to avoid hitting the rate limit. 🚦 How do rate limits work?CodeRabbit enforces hourly rate limits for each developer per organization. Our paid plans have higher rate limits than the trial, open-source and free plans. In all cases, we re-allow further reviews after a brief timeout. Please see our FAQ for further information. ℹ️ Review info⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (5)
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
broomva
added a commit
that referenced
this pull request
May 12, 2026
…l engine (#8) Bstack now treats broomva.tech/prompts as a shared knowledge surface across agents — not a private repository. Every bstack-driven workspace gets two affordances from the surface: 1. Reusable directives (code review, deep research, harness bootstrap, platform redesign, etc.) — versioned, parameterized, ranked by usage. 2. An evaluation engine that writes a typed prompt_invocation row for every pull and every completion (source attribution, latency, tokens, cost, feedback). Mirrors the auto-tracing mandate from the broomva-cli skill SKILL.md so the bstack agent reading this file knows it owes a completion call after every pull. Changes: - NEW references/prompts-integration.md (201 lines) — canonical guide: when to reach for the library, discovery patterns, the 5-step mandate (set source → pull --json → use body → complete with model/tokens → optional feedback), source attribution table, env vars, composition with primitives P1/P4/P6/P11/P13, and common traps. - SKILL.md — new "Prompts as a shared knowledge surface" section with the reflexive mandate inline (so an agent loading SKILL.md alone sees the contract) and a pointer to the full reference file. See-also list extended. - references/skills-roster.md — refreshed row #6 description from the stale "Pull, push, update, delete via CLI" to the current Phase 1+2 reality (eval engine, source attribution, Rust CLI, auto-tracing mandate). - references/quickstart.md — Memory row of the layer table updated to reference the shared knowledge surface + eval engine, with a more natural trigger phrase ("use the code-review-agent prompt on this diff"). Co-authored-by: Test <test@test.com> Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
5 tasks
broomva
pushed a commit
that referenced
this pull request
May 22, 2026
…3 follow-up) Promotes the LLM-as-index /kg load skill from workspace-local v1 (shipped in 0.12.0 at ~/.claude/skills/kg/) to a managed bstack companion skill. Closes the "Future work" item explicitly anchored in the 0.12.0 CHANGELOG: promote broomva/kg to its own GitHub repo + bstack skills-lock entry once usage exceeds rule-of-three. Repo is live at https://github.com/broomva/kg (MIT-licensed; SKILL.md SHA256 1cf125beaeb675580cc18ef3007ca92eff734db0dc16467978a6d896aa5657a7). - CHANGED references/companion-skills.yaml — adds `kg` as a required, knowledge-category, P6 skill placed next to knowledge-graph-memory (its sibling in the P6 family: load-end vs capture-end of the bookkeeping loop). bin/bstack-skills install picks it up automatically via parse_roster() — no install-script changes needed. - CHANGED scripts/doctor.sh §7 — /kg load check now accepts either a managed install (~/.claude/skills/kg/ OR ~/.agents/skills/kg/) or a legacy workspace-local v1 install. Gap message points at `npx skills add broomva/kg` (the new managed path). - CHANGED references/skills-roster.md — adds row #6 in Memory & Consciousness, between knowledge-graph-memory and prompt-library; full table renumbered (continuous 1-28); header count 27 → 28. - CHANGED VERSION — 0.13.0 → 0.14.0. Minor bump: additive but the required roster surface area changes (existing v0.13.0 installs will see kg as "missing*" in `bstack skills status` until `bstack skills install` runs). Schema unchanged — new entry validates against companion-skills.v1.json. tests/skills-roster.test.sh Test 1 GREEN with 32 skills (was 31). bstack doctor post-change: 86/88 (identical to pre-change baseline; the 2 gaps are pre-existing P19 mechanism + P6 catalog hook concerns about the worktree, unrelated to kg). Why P6, not a new primitive: kg operationalizes P6 at the load end of the loop (dual of knowledge-graph-memory which captures into the graph). Same primitive, opposite arrow. No L3 stability-budget impact. Refs: BRO-1223 Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
broomva
added a commit
that referenced
this pull request
May 22, 2026
…1223 follow-up) (#44) * feat(0.14.0): register broomva/kg as managed companion skill (BRO-1223 follow-up) Promotes the LLM-as-index /kg load skill from workspace-local v1 (shipped in 0.12.0 at ~/.claude/skills/kg/) to a managed bstack companion skill. Closes the "Future work" item explicitly anchored in the 0.12.0 CHANGELOG: promote broomva/kg to its own GitHub repo + bstack skills-lock entry once usage exceeds rule-of-three. Repo is live at https://github.com/broomva/kg (MIT-licensed; SKILL.md SHA256 1cf125beaeb675580cc18ef3007ca92eff734db0dc16467978a6d896aa5657a7). - CHANGED references/companion-skills.yaml — adds `kg` as a required, knowledge-category, P6 skill placed next to knowledge-graph-memory (its sibling in the P6 family: load-end vs capture-end of the bookkeeping loop). bin/bstack-skills install picks it up automatically via parse_roster() — no install-script changes needed. - CHANGED scripts/doctor.sh §7 — /kg load check now accepts either a managed install (~/.claude/skills/kg/ OR ~/.agents/skills/kg/) or a legacy workspace-local v1 install. Gap message points at `npx skills add broomva/kg` (the new managed path). - CHANGED references/skills-roster.md — adds row #6 in Memory & Consciousness, between knowledge-graph-memory and prompt-library; full table renumbered (continuous 1-28); header count 27 → 28. - CHANGED VERSION — 0.13.0 → 0.14.0. Minor bump: additive but the required roster surface area changes (existing v0.13.0 installs will see kg as "missing*" in `bstack skills status` until `bstack skills install` runs). Schema unchanged — new entry validates against companion-skills.v1.json. tests/skills-roster.test.sh Test 1 GREEN with 32 skills (was 31). bstack doctor post-change: 86/88 (identical to pre-change baseline; the 2 gaps are pre-existing P19 mechanism + P6 catalog hook concerns about the worktree, unrelated to kg). Why P6, not a new primitive: kg operationalizes P6 at the load end of the loop (dual of knowledge-graph-memory which captures into the graph). Same primitive, opposite arrow. No L3 stability-budget impact. Refs: BRO-1223 Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * feat(doctor): catalog stale threshold reads from .control/policy.yaml §7 P6 catalog freshness check now sources its threshold from .control/policy.yaml catalog.stale_doctor_hours, with a hardcoded 48h fallback when: - policy.yaml is missing - python3 is unavailable - PyYAML is missing (read fails) This is part of the BRO-1223 follow-up to reconcile three different "stale" thresholds previously hardcoded across: - ~/.claude/skills/kg/scripts/kg.py (24h, now policy.stale_warn_hours) - ~/broomva/bstack/scripts/doctor.sh (48h, now policy.stale_doctor_hours) - ~/broomva/scripts/knowledge-catalog-refresh-hook.sh (300s, now policy.refresh_cooldown_seconds) After this PR + the paired broomva/workspace PR, policy.yaml is the single source of truth. Doctor reports the active threshold in the OK message so the user sees which threshold applied. Pairs with broomva/workspace PR (adds the catalog: block to policy.yaml and updates kg.py + hook to read from it). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * fix(doctor): P20 round-1 — argv passing + regex validation + honest CHANGELOG Addresses cross-review verdict 6/10 FAIL: - **C1 (critical)**: CHANGELOG 0.14.0 previously described only commit a261461 (kg registration). Now also describes commit 8de2f1b's doctor.sh §7 policy.yaml threshold reader — which is the primary BRO-1223 I1 closure shipping in this PR. The CHANGELOG is the audience-facing projection of the diff; honesty about what shipped is non-negotiable. - **C3 (critical)**: doctor.sh §7 `_stale_h` now regex-validates the python3 output (`[[ "$_raw" =~ ^[0-9]+$ ]]`). Empty stdout from a malformed policy.yaml previously corrupted the integer comparison and falsely flagged a stale catalog every run. - **I1 (important)**: doctor.sh §7 python3 invocation now uses argv-passed args via sys.argv (avoids SyntaxError on single-quote-in-path edge case). Same hardening pattern as the paired workspace hook. Pairs with the workspace PR (kg.py bounds check + bench KG_PY + hook argv/regex hardening). Refs BRO-1223 P20 round-1. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Carlos D. Escobar-Valbuena <devteam@getstimulus.ai> Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Companion to broomva/workspace#45 (P13 governance). Extends doctor 12→13. Closes a doc gap where references/primitives.md was missing P12 entirely — now covers P12 + P13 with reflexive triggers + cohesion narrative + rule-of-three deferral note for morpheus. Smoke-tested 46/49 against post-merge target.