Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/Mark393295827/third-brain-v7-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/rules/mark393295827/third-brain-v7-skills/third-brain-skills)<a href="https://agentmods.dev/rules/mark393295827/third-brain-v7-skills/third-brain-skills"><img src="https://agentmods.dev/badge/rules/mark393295827/third-brain-v7-skills/third-brain-skills/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/rules/mark393295827/third-brain-v7-skills/third-brain-skills"><img src="https://agentmods.dev/badge/rules/mark393295827/third-brain-v7-skills/third-brain-skills.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00016 | $0.00753 |
| Opus 5 | $0.00008 | $0.00377 |
| Sonnet 5 | $0.00003 | $0.00151 |
| Haiku 4.5 | $0.00002 | $0.00075 |
Grade A, and why
third-brain-skills scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Third Brain V8.1 Skill Router
When the user asks for knowledge ingestion, daily review, research, verification, behavior design, creativity, startup evaluation, or agent-team orchestration:
- Inspect the relevant
skills/<skill-name>/SKILL.md. - Read
metadata.profile,assumes, andconflicts_withbefore acting. - Follow
Usage TemplateandWorkflow, including the unknowns gate and evaluation. - Emit only the standard failure codes defined by the skill.
- Return the skill's Output Contract with verification evidence and residual risk.
For wiki-writing skills, resolve paths from system/config.md when present. Use defaults only when no project config exists.
Common routing:
- Source, PDF, article, or note ingestion ->
skills/wiki-ingest/SKILL.md - Daily review or daily planning ->
skills/daily-okr/SKILL.md - Deep understanding or synthesis ->
skills/cognitive-compile/SKILL.md - Completion claim, bug fix, tests, shipping ->
skills/verify-before-claim/SKILL.md - Habit or behavior change ->
skills/behavior-design/SKILL.md - Idea generation ->
skills/creativity-engine/SKILL.md - Research report, market scan, competitor analysis, or fast-changing evidence question ->
skills/deep-research/SKILL.md - Startup idea or due diligence ->
skills/startup-evaluation/SKILL.md - Property operations, dispatch, quotes, evidence, CTQ metrics, or AI Six Sigma ->
skills/ai-six-sigma-property-os/SKILL.md - Knowledge organization, deduplication, or vector search ->
skills/knowledge-ops/SKILL.md - Wiki health check, broken links, or stale pages ->
skills/wiki-lint/SKILL.md - Context limits or token budget ->
skills/context-manager/SKILL.md - Project triage, WIP, blockers, or completion state ->
skills/project-flow-ops/SKILL.md - Session closure or reusable learning extraction ->
skills/session-learn/SKILL.md - Repeatable task, scheduled routine, or bounded correction loop ->
skills/loop-engineering/SKILL.md - Explicit dependencies, independently executable branches, typed joins, or node-local recovery ->
skills/graph-engineering/SKILL.md - Agent workflow refactor or autonomy design ->
skills/agentic-engineering/SKILL.md - Agent permissions, tools, observability, or runtime safety ->
skills/harness-engineering/SKILL.md - Team operating system, growth method, or self-evolving work method ->
skills/anthropic-os/SKILL.md - Multi-agent execution ->
skills/agent-teams-command/SKILL.md - Audit manual work, prior sessions, or interviews for reusable workflows ->
skills/workflow-audit/SKILL.md
Engineering boundary: Loop = temporal depth; Graph = dependency width; Agent Teams = process ownership, IPC, and integration; Harness = runtime scheduler, permissions, and observability. Use graph-engineering only when admission value exceeds orchestration and review cost. V8.1 supports bounded static DAGs, not dynamic or cyclic graphs.
Do not claim completion unless fresh evidence satisfies Success Metrics and Quality Gates.
For high-risk profiles, require independent verification, approval boundaries, rollback, and an audit receipt. For stateful, loop, and high-risk profiles, persist state before returning.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 10d ago First seen · 47 lines · 16 tokens per session scan A ab9a06f4cd23
third-brain-skills is a cursor rule published in the GitHub repository Mark393295827/third-brain-v7-skills (138 stars, last pushed 19d ago), licensed MIT. It adds 16 tokens to every session and 753 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other cursor rules, from other repositories
nextjs
Rules for Next.js projects (App Router and Pages Router).
php
Rules applied when working in PHP files (incl. Laravel, Symfony, WordPress).
shell
Rules for shell scripts (bash, zsh, sh, PowerShell).
vue
Rules for Vue (2 and 3) single-file components.
tests
Rules for test files — keep tests lean and pattern-matched.
ponytail
Ponytail, lazy senior dev mode. Always pick the simplest solution that works.