Skip to content
Back to search

third-brain-v7-skills

agent wiki +engineering skills

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/mark393295827--third-brain-v7-skills/badge.svg)](/listings/mark393295827--third-brain-v7-skills)

HTML

<a href="/listings/mark393295827--third-brain-v7-skills"><img src="/api/listings/mark393295827--third-brain-v7-skills/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

1. Clone & Install

git clone https://github.com/Mark393295827/third-brain-v5-skills.git
cd third-brain-v5-skills
bash install.sh codex  # Codex is the primary host/kernel

Windows PowerShell uses `.\install.ps1 codex`. Compatibility adapters require an explicit non-Codex target.

2. Try a Skill

Paste this into Codex:

Use wiki-ingest on this source. Stage an immutable source, author a governed concept candidate, plan navigation updates, and return a verification summary. Treat entity authoring as a separate reviewed capability.

Then paste a URL, article, PDF text, or any source you want to capture.

3. Next Steps

See the **[3-Minute Quickstart](examples/3-minute-quickstart.md)** for a complete walkthrough.

Full guide: **[GUIDE.md](GUIDE.md)**

Command centre and distribution: **[V8.1 Codex Agentic OS command centre](docs/agentic-os-command-center.md)**. Start it with `python tools/agentic_os_server.py --state-root .agentic-os-state`; the host is localhost-only by default and requires T02 runtime receipts. Claude/Gemini/Cursor/Windsurf remain compatibility adapters, not the primary runtime.

中文历史进阶手册:**[V7 最大潜力使用手册(参考)](docs/v7-max-potential-guide-zh.md)**

🔥 **[V8.1 多智能体 Worker 流水线手册](docs/worker-flows-guide-zh.md)** & **[Worker Flows 规范](workflows/worker-flows.md)**:事务式流水线(Stage → Author → Graph Plan → Govern → Commit/Postcheck → optional Deliverable),归档只发生在验证提交之后。

---

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 380 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # Third Brain V8.1 Skills
  2. <p align="center">
  3. <img src="assets/third-brain-v8.1-actual-system-architecture.png" alt="Third Brain V8.1 actual Codex OS architecture: Codex host, repository control plane, transactional worker runtime, live Obsidian Vault, verification evidence, and declared debt" width="900">
  4. </p>
  5. **Verification-first Agent Skills for Codex OS, with optional compatibility adapters for Claude Code, Gemini, Cursor, and Windsurf.** Build a persistent knowledge operating system with Obsidian provenance, scheduled loops, bounded static dependency graphs, context management, and multi-agent orchestration.
  6. Install 21 reusable agent skills for ingesting sources, auditing workflows, compiling interlinked wikis, running daily knowledge loops, verifying claims before shipping, managing token costs, engineering bounded loops and static dependency graphs, and orchestrating multi-agent teams.
  7. **Use this if:**
  8. - Your AI agent keeps forgetting context between sessions
  9. - You want to enforce verification gates before "done" claims
  10. - You need to structure knowledge as linked pages (like Obsidian) instead of scattered chat history
  11. - You're building agent workflows that should learn and compound over time
  12. - You want Obsidian wiki knowledge to update skills, SOPs, schemas, and automation through supervised promotion gates
  13. [![GitHub stars](https://img.shields.io/github/stars/Mark393295827/third-brain-v5-skills?style=social)](https://github.com/Mark393295827/third-brain-v5-skills/stargazers)
  14. [![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](LICENSE)
  15. [![Codex OS](https://img.shields.io/badge/Codex%20OS-Primary-10A37F)](AGENTS.md)
  16. [![Claude Code](https://img.shields.io/badge/Claude%20Code-Compatibility%20Adapter-8A2BE2)](docs/compatibility.md)
  17. [![Cursor](https://img.shields.io/badge/Cursor-Rules%20Compatible-111827)](GUIDE.md)
  18. [![PRs Welcome](https://img.shields.io/badge/PRs-welcome-brightgreen)](CONTRIBUTING.md)
  19. ---
  20. ## Quick Start
  21. ### 1. Clone & Install
  22. ```bash
  23. git clone https://github.com/Mark393295827/third-brain-v5-skills.git
  24. cd third-brain-v5-skills
  25. bash install.sh codex # Codex is the primary host/kernel
  26. ```
  27. Windows PowerShell uses `.\install.ps1 codex`. Compatibility adapters require an explicit non-Codex target.
  28. ### 2. Try a Skill
  29. Paste this into Codex:
  30. ```text
  31. Use wiki-ingest on this source. Stage an immutable source, author a governed concept candidate, plan navigation updates, and return a verification summary. Treat entity authoring as a separate reviewed capability.
  32. ```
  33. Then paste a URL, article, PDF text, or any source you want to capture.
  34. ### 3. Next Steps
  35. See the **[3-Minute Quickstart](examples/3-minute-quickstart.md)** for a complete walkthrough.
  36. Full guide: **[GUIDE.md](GUIDE.md)**
  37. Command centre and distribution: **[V8.1 Codex Agentic OS command centre](docs/agentic-os-command-center.md)**. Start it with `python tools/agentic_os_server.py --state-root .agentic-os-state`; the host is localhost-only by default and requires T02 runtime receipts. Claude/Gemini/Cursor/Windsurf remain compatibility adapters, not the primary runtime.
  38. 中文历史进阶手册:**[V7 最大潜力使用手册(参考)](docs/v7-max-potential-guide-zh.md)**
  39. 🔥 **[V8.1 多智能体 Worker 流水线手册](docs/worker-flows-guide-zh.md)** & **[Worker Flows 规范](workflows/worker-flows.md)**:事务式流水线(Stage → Author → Graph Plan → Govern → Commit/Postcheck → optional Deliverable),归档只发生在验证提交之后。
  40. ---
  41. ## The Problem
  42. | Scenario | Before | After |
  43. |----------|--------|-------|
  44. | **Research PDF** | Summarized once, then forgotten | `wiki-ingest`: source notes, concept pages, linked wiki |
  45. | **Coding session** | Agent: "I fixed it" (no proof) | `verify-before-claim`: requires test output + exit code before any claim |
  46. | **Daily work** | Tasks and ideas scatter across chat | `daily-okr`: one insight, one wiki update, one action, one output, daily score |
  47. ---
  48. ## V8.1 Operating Model
  49. V8.1 keeps the wiki as durable disk and governance, gives every skill a profile-aware execution contract, and makes `python -m tools.worker_flow.cli` the only canonical transactional runtime:
  50. ```text
  51. Input -> Source -> Wiki compile -> Daily loop -> Agent/Wiki flywheel -> Skill/SOP upgrade -> Verification
  52. ```
  53. The upgrade adds seven hard defaults:
  54. - **Profiles match risk**: one-shot, stateful, loop, and high-risk work load only the controls they need.
  55. - **Source provenance stays immutable**: raw source notes and block refs remain the audit layer.
  56. - **Loops have contracts**: Trigger -> Execute -> Verify -> State, with budgets and recovery.
  57. - **Context is zero-overhead by default**: hot paths load only what the task needs.
  58. - **Automation is bounded**: scans and queues can run unattended; semantic writes stay supervised.
  59. - **Teams need ownership**: multi-agent work requires write scopes, IPC, join gates, cleanup, and evidence.
  60. - **Rules promote through evidence**: wiki insights become schema or skill rules only after repeated support and a cheap check.
  61. See the [V8.1 Worker Flow](workflows/worker-flows.md), the [Agent Skills standard](docs/agent-skills-standard.md), and the historical [V7 release notes](docs/release-notes-v7.md).
  62. V8.1 keeps `graph-engineering` bounded to static DAGs; dynamic expansion and cyclic graphs remain out of scope.
  63. ---
  64. ## Core Skills (Start Here)
  65. ### 🧠 **Knowledge & Verification**
  66. | Skill | What it does |
  67. |-------|-------------|
  68. | **[wiki-ingest](skills/wiki-ingest/SKILL.md)** | Preserve an immutable source, compile linked understanding, enforce promotion boundaries, and verify every touched file. |
  69. | **[verify-before-claim](skills/verify-before-claim/SKILL.md)** | **Iron rule**: no completion claim without fresh, scope-matched evidence; consequential actions also require independent verification, approval, and rollback. |
  70. ### 📅 **Daily Loop**
  71. | Skill | What it does |
  72. |-------|-------------|
  73. | **[daily-okr](skills/daily-okr/SKILL.md)** | Execute a 7-KR cycle: Input → Cognition → Wiki → Behavior → Creativity → Output → Feedback, with an artifact or receipt for every completed KR. |
  74. ### 🎯 **Behavior & Creativity**
  75. | Skill | What it does |
  76. |-------|-------------|
  77. | **[behavior-design](skills/behavior-design/SKILL.md)** | Turn outcomes into observable minimum behavior, cues, SOPs, recovery, evidence, and review. |
  78. | **[creativity-engine](skills/creativity-engine/SKILL.md)** | Generate mechanism-diverse options, rank evidence gaps, and define three bounded minimum experiments. |
  79. ### 🔬 **Research & Quality**
  80. | Skill | What it does |
  81. |-------|-------------|
  82. | **[deep-research](skills/deep-research/SKILL.md)** | STOW-compatible research harness with evidence trails, source/claim ledgers, privacy checks, and direct handoff to wiki-ingest. |
  83. | **[session-learn](skills/session-learn/SKILL.md)** | Extract knowledge patterns from sessions — concepts, entities, corrections, patterns, ideas, decisions, gaps. Closure protocol ensures learning feeds back into the wiki. |
  84. ### 📊 **Context & Engineering**
  85. | Skill | What it does |
  86. |-------|-------------|
  87. | **[context-manager](skills/context-manager/SKILL.md)** | Derive budgets at runtime, checkpoint state, compact with KEEP/SUMMARIZE/DROP/RETRIEVE, and route by capability. |
  88. | **[loop-engineering](skills/loop-engineering/SKILL.md)** | Control temporal depth: turn repeatable work into a bounded Trigger -> Execute -> Verify -> State loop with durable state, finite budgets, and explicit stop/recovery rules. |
  89. | **[graph-engineering](skills/graph-engineering/SKILL.md)** | Control dependency width: validate bounded static DAGs with explicit dependencies, independent branches, typed joins, and node-local recovery. |
  90. | **[agentic-engineering](skills/agentic-engineering/SKILL.md)** | Design agent workflows as spec-driven macro actions with quality ceilings, verification gates, delegated-action boundaries, and state checkpoints. |
  91. | **[harness-engineering](skills/harness-engineering/SKILL.md)** | Provide the runtime scheduler, permissions, leases, tools, provenance, and observability needed to execute agent workflows safely. |
  92. | **[agent-teams-command](skills/agent-teams-command/SKILL.md)** | Admit and command multi-agent processes with exclusive ownership, typed IPC, isolated writes, serial integration, evidence, and cleanup. |
  93. ---
  94. ## Complete Skill List
  95. ### 📥 Ingestion & Knowledge
  96. - `wiki-ingest` — Stage immutable sources, author governed concept candidates, plan graph updates, and hand off optional entity authoring for separate review
  97. - `knowledge-ops` — Multi-layer knowledge management: classify, deduplicate, preserve evidence hierarchy, vector + Markdown retrieval
  98. - `wiki-lint` — Health check: P0/P1 graph health, frontmatter, source refs, wikilink density, provenance debt
  99. ### 🔄 Daily Workflow
  100. - `daily-okr` — 7-KR evidence cycle with minimum-day degradation and durable closeout
  101. - `cognitive-compile` — 8-section framework: Question → Facts → Concepts → Pattern Recognition → Conflict Detection → Hypothesis Generation
  102. ### 🎨 Behavior & Creativity
  103. - `behavior-design` — Outcome → minimum behavior → cue → SOP → evidence → review
  104. - `creativity-engine` — Mechanism-diverse combinations, scoring, and falsifiable minimum tests
  105. ### 🔬 Research & Quality
  106. - `deep-research` — STOW-compatible research harness, evidence trails, source/claim ledgers
  107. - `verify-before-claim` — Fresh scope-matched evidence, independent checks, approval, rollback
  108. ### 🔄 Learning & Flow
  109. - `session-learn` — 7 signal types: concepts, entities, corrections, patterns, ideas, decisions, gaps
  110. - `project-flow-ops` — Execution flow: triage, plan, track, review across projects
  111. ### 📊 Context
  112. - `context-manager` — Runtime budgets, checkpoint replay, compaction, retrieval, capability routing
  113. Utility command: `commands/token-cost-tracker.md` estimates, logs, and reports token usage; it is not one of the 21 Agent Skills.
  114. ### 🏗️ Engineering
  115. - `loop-engineering` — Temporal depth through bounded loop contracts, independent verification, finite budgets, and stop/recovery rules
  116. - `graph-engineering` — Dependency width through bounded static DAGs, explicit branches, typed joins, and node-local recovery
  117. - `agentic-engineering` — Spec-driven macro actions, quality ceilings, verification gates, state checks
  118. - `harness-engineering` — Runtime scheduler, permissions, system-call tools, delegated gates, provenance, observability
  119. - `agent-teams-command` — Multi-agent process ownership, IPC, integration, async budgets, and evidence gates
  120. ### 💼 Strategy & Operations
  121. - `startup-evaluation` — Startup health: entrepreneurship, VC 5T, PMF, runway, team, unit economics
  122. - `anthropic-os` — Governed operating-system experiments: Four-C, 3B, prediction errors, permission ladder
  123. - `ai-six-sigma-property-os` — AI + Ontology + DMAIC for property work orders, dispatch, quotes, evidence
  124. ---
  125. ## How These Skills Fit Together
  126. ```
  127. LLM (CPU) + Context (RAM) + Wiki (Disk)
  128. External Sources ──→ wiki-ingest + knowledge-ops ──→ Knowledge Layers
  129. ↓
  130. Daily Loop (daily-okr)
  131. / / \ \ \
  132. Input Cognition Wiki Behavior Creativity
  133. |
  134. Output → Feedback
  135. ↓
  136. session-learn (extract patterns)
  137. ↓
  138. verify-before-claim (quality gate)
  139. ↓
  140. Loop depth / Graph dependency width
  141. ↓
  142. Multi-agent teams (agent-teams-command)
  143. ```
  144. The system is a **closed loop**: ingest sources → process daily → extract learning → verify claims → promote rules → scale to teams.
  145. ---
  146. ## Architecture & Design
  147. **Third Brain V8.1 treats agents as LLM OS processes:**
  148. - LLM = CPU
  149. - Context = RAM
  150. - Wiki/Obsidian = Disk
  151. - Tools = System calls
  152. - Skills = Executable programs
  153. - Harness = Kernel
  154. - Agent teams = Processes
  155. ### Design Layers
  156. | Layer | Principle | Skills |
  157. |-------|-----------|--------|
  158. | **🧠 Knowledge OS** | Capture, structure, lint, and promote knowledge over time | wiki-ingest, knowledge-ops, wiki-lint |
  159. | **⚡ Daily Loop** | Close the knowledge-to-action cycle every day | daily-okr, cognitive-compile |
  160. | **🎯 Behavior & Creativity** | Turn knowledge into habits and novel ideas | behavior-design, creativity-engine |
  161. | **🔬 Research & Quality** | Verify before claiming, research with rigor | deep-research, verify-before-claim |
  162. | **🔄 Continuous Learning** | Extract patterns from every session | session-learn, project-flow-ops |
  163. | **📊 Context** | Manage hot context, durable checkpoints, and runtime budgets | context-manager |
  164. | **🏗️ Engineering** | Design bounded loops, static dependency graphs, harnesses, agent workflows, and multi-agent systems | loop-engineering, graph-engineering, agentic-engineering, harness-engineering, agent-teams-command |
  165. | **💼 Strategy & Operations** | Evaluate startups, design AI quality systems | startup-evaluation, anthropic-os, ai-six-sigma-property-os |
  166. Engineering boundaries are deliberate: Loop Engineering controls temporal depth; Graph Engineering controls dependency width; Agent Teams Command controls process ownership, IPC, and integration; Harness Engineering supplies the runtime scheduler, permissions, and observability. A graph node may contain a Loop or Agent Team, but V8.1 does not support dynamic or cyclic graphs.
  167. ---
  168. ## Skill Adoption Path
  169. Start small. Add skills as you need them.
  170. | Stage | Core Skills | Unlock When |
  171. |-------|------------|-------------|
  172. | **Week 1** | `wiki-ingest` + `verify-before-claim` | You can ingest 1 source/day + every claim has fresh evidence |
  173. | **Weeks 2-4** | + `daily-okr` + `session-learn` | Daily artifacts and verified learnings feed back to the wiki |
  174. | **Month 2+** | + `cognitive-compile` + `behavior-design` + `creativity-engine` | Wiki has 50+ pages or repeated decisions need synthesis |
  175. | **Month 3+** | + `knowledge-ops` + `loop-engineering` + `harness-engineering` + `agentic-engineering` | Retrieval, looping reliability, permissions, or delegated actions become bottlenecks |
  176. | **Dependency graph** | + `graph-engineering` after `loop-engineering` | Explicit dependencies, independent branches, typed joins, or node-local recovery create more value than orchestration and review cost |
  177. | **Multi-agent** | + `agent-teams-command` + `project-flow-ops` | Work splits into separate owners with clear integration gates |
  178. | **Strategy** | + `startup-evaluation` + `anthropic-os` + `deep-research` | Need startup health, market, operating-system decisions |
  179. | **Operations** | + `ai-six-sigma-property-os` | Need measurable service quality, dispatch, evidence loops |
  180. ---
  181. ## Installation
  182. ### For Codex CLI (Primary host/kernel)
  183. ```bash
  184. # Personal skills (available across all projects)
  185. bash install.sh codex
  186. ```
  187. ### For Cursor
  188. ```bash
  189. bash install.sh cursor
  190. ```
  191. ### For Windsurf
  192. ```bash
  193. bash install.sh windsurf
  194. ```
  195. ### Compatibility adapters: Claude Code / Gemini CLI
  196. ```bash
  197. bash install.sh claude
  198. bash install.sh gemini
  199. ```
  200. Full guide: **[GUIDE.md](GUIDE.md)**
  201. ---
  202. ## Example Workflows
  203. Each workflow is copyable into your agent. See **[examples/](examples/)** for complete, verified examples:
  204. - [3-minute quickstart](examples/3-minute-quickstart.md) — Fastest path to useful output
  205. - [Research PDF to wiki](examples/research-pdf-to-wiki.md) — Turn a source into linked pages
  206. - [Verified code session](examples/verified-code-session.md) — Use verification gates before claiming a fix
  207. - [Daily knowledge loop](examples/daily-knowledge-loop.md) — Run a compact daily OKR workflow
  208. - [Startup evaluation sprint](examples/startup-evaluation-sprint.md) — Turn an idea into validated assumptions
  209. ---
  210. ## Wiki Structure
  211. Skills default to STOW paths (configurable via `system/config.md`):
  212. ```
  213. sources/ ← Immutable source notes (articles, PDFs, transcripts)
  214. wiki/
  215. ├── concepts/ ← 13 contracted domains
  216. ├── entities/ ← people, companies, funds-investors, products, orgs
  217. └── outputs/ ← gmail-digests, evaluations, compilations
  218. maps/
  219. ├── domain-mocs/
  220. ├── system-indexes/
  221. ├── project-maps/
  222. └── canvases/
  223. system/ ← Deployed contracts, config, schema, templates, runs
  224. ```
  225. ---
  226. ## Tools & References
  227. | Resource | Purpose |
  228. |----------|---------|
  229. | [tools/index.html](tools/index.html) | V8.1 visual skill navigator |
  230. | [tools/token-calculator.html](tools/token-calculator.html) | Historical V5 cost-model calculator; not a runtime authority |
  231. | [GUIDE.md](GUIDE.md) | Full installation & troubleshooting |
  232. | [V7 最大潜力使用手册(历史参考)](docs/v7-max-potential-guide-zh.md) | Earlier profile、知识飞轮、Agent 工程、治理和 30 天采用路径 |
  233. | [CLAUDE.md](CLAUDE.md) | Claude Code setup |
  234. | [CONTRIBUTING.md](CONTRIBUTING.md) | How to contribute skills |
  235. ---
  236. ## What This Is NOT
  237. - **Not a chat wrapper.** Skills are executable prompts that agents follow; they're not productivity theater.
  238. - **Not a productivity tool with 100 metrics.** Daily OKR has seven causally linked KRs with evidence.
  239. - **Not an all-in-one framework.** Pick skills incrementally. You don't need all 20 to start.
  240. - **Not prescriptive.** Adapt paths, frontmatter, and skill triggers to your workflow.
  241. ---
  242. ## Philosophy
  243. **Three core principles:**
  244. 1. **Verification first**: No "done" without proof. No claims without evidence. Expected value over confidence.
  245. 2. **Knowledge compounds**: Every session should improve the wiki, not scatter across chat history.
  246. 3. **Closed loops**: Ingest → Process → Learn → Scale → Verify. No loose ends.
  247. ---
  248. ## Related Work
  249. - [Agent Skills Format](https://agentskills.io) — Open specification for agent skills
  250. - [llm-wiki-agent](https://github.com/SamurAIGPT/llm-wiki-agent) — Original STOW pattern implementation by SamurAIGPT
  251. - [Karpathy LLM OS](https://karpathy.ai) — Conceptual framework
  252. ---
  253. ## Contributing
  254. Bug reports, skill improvements, and PRs are welcome. See [CONTRIBUTING.md](CONTRIBUTING.md) and [CHANGELOG.md](CHANGELOG.md).
  255. ---
  256. ## License
  257. MIT — see [LICENSE](LICENSE).
  258. ---
  259. ## Transparency Note
  260. This project includes **growth and outreach tools** (in `tools/` and `outreach/`) designed to help the repository reach users via GitHub search and Awesome lists. These tools are optional, dry-run-by-default, and separate from the core skills. If you're installing skills for your own workflow, you don't need them.
  261. The skill frameworks and philosophy are genuine. Use what works for you; ignore what doesn't.

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.