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.
npx agentmods add rules/huytieu/cog-second-brain/cursorrulesgit clone --depth 1 https://github.com/huytieu/COG-second-brainWhat 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 | $0.00732 | $0.00732 |
| Opus 5 | $0.00366 | $0.00366 |
| Sonnet 5 | $0.00146 | $0.00146 |
| Haiku 4.5 | $0.00073 | $0.00073 |
Grade A, and why
cursorrules 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 2d 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.
How it starts
The opening of the file, as written. The whole thing — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
COG Second Brain — Cursor Rules
This is the Cursor-compatible version of CLAUDE.md
COG: Cognition + Obsidian + Git — a self-evolving agentic second brain
Inspired by Garry Tan's gstack and gbrain patterns
Skills
COG exposes 17 skills via .claude/skills/*/SKILL.md. Each skill has a name and description in YAML frontmatter. Use AGENTS.md as the universal command reference.
Worker Agents
10 agents live in .claude/agents/: 6 workers handle data-heavy tasks, 4 read-only verifiers check the result against acceptance criteria by observing the artifact. The lead session handles reasoning and synthesis.
- worker-data-collector: Structured extraction from GitHub, Slack, Jira, Linear
- worker-researcher: Web research with source citations
- worker-file-ops: Vault file operations, metadata, profiles
- worker-executor: Pre-approved mutations (Jira, Linear, APIs)
- worker-publisher: Publishing to Slack, Confluence, Notion
- brief-people-updater: Batch-update people profiles
Brain-First Protocol
Before answering questions about people, projects, strategy, or decisions:
- Read relevant notes from
05-knowledge/first - If project-specific, read
04-projects/<project>/ - Then synthesize an answer
Daily Journal
Ambient, not a command. After finishing a meaningful unit of work (shipped or committed something, produced a deliverable, made a decision, hit a blocker), append one entry to 01-daily/journal/YYYY-MM-DD.md (date +%F). Create the file from the template in .claude/skills/daily-journal/SKILL.md on the first entry of the day; append only. Skip trivia and scratch work, and do not announce the write.
Integration Preferences
Check 00-inbox/MY-INTEGRATIONS.md before using any external integration:
- Active: use normally
- Disabled: skip silently
- Unknown: ask the user
Vault Structure
00-inbox/ — Profiles, interests, integrations
01-daily/ — Briefs and check-ins
02-personal/ — Personal braindumps (private)
03-professional/ — Professional braindumps and strategy
04-projects/ — Per-project tracking
05-knowledge/ — Consolidated insights, people CRM, patterns
06-templates/ — Document templates
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.
- 2d ago First seen · 59 lines · 732 tokens per session scan A be3118fbb97b
cursorrules is a cursor rule published in the GitHub repository huytieu/COG-second-brain (1,153 stars, last pushed 7d ago), licensed MIT. It adds 732 tokens to every session, about $0.0037 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
orchestrator
Use when coordinating AI agents through a pipeline — orchestrator role instructions.
workspace
Multi-agent workspace context — loaded on every session.
engineering-agent-prompt-engineer
Use when writing, reviewing, or improving prompts for AI agents — agent role design and prompt engineering.
rules
Agentic Playwright orchestrator -- constitution, workflow, and skills index.
dev_workflow
Guide for using Taskmaster to manage task-driven development workflows.
fix-issue
Implement a fix following the human-thinking loop — understand the root cause, plan the minimal change, implement, verify the problem is actually gone.