huytieu/COG-second-brain
Self-evolving second brain with 35 AI skills, 10 agents, and people CRM. Closed-loop harness: a V-model verification lifecycle where the worker never grades its own homework. Plus paired anti-slop design skills for marketing and product UI.
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
Copy this rig
# review before running: this installs third-party code
$ npx degit huytieu/COG-second-brain/.claude ./rig-cog-second-brain # inspect, then merge into .claude/ MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.
$ npx degit huytieu/COG-second-brain/.agents/skills/auto-research .claude/skills/auto-research $ npx degit huytieu/COG-second-brain/.agents/skills/braindump .claude/skills/braindump $ npx degit huytieu/COG-second-brain/.agents/skills/closed-loop .claude/skills/closed-loop $ npx degit huytieu/COG-second-brain/.agents/skills/comprehensive-analysis .claude/skills/comprehensive-analysis $ npx degit huytieu/COG-second-brain/.agents/skills/content-factory .claude/skills/content-factory $ npx degit huytieu/COG-second-brain/.agents/skills/create-user-story .claude/skills/create-user-story $ npx degit huytieu/COG-second-brain/.agents/skills/daily-brief .claude/skills/daily-brief $ npx degit huytieu/COG-second-brain/.agents/skills/daily-journal .claude/skills/daily-journal $ npx degit huytieu/COG-second-brain/.agents/skills/data-forms .claude/skills/data-forms $ npx degit huytieu/COG-second-brain/.agents/skills/editorial-illustrations .claude/skills/editorial-illustrations $ npx degit huytieu/COG-second-brain/.agents/skills/export-open-issues .claude/skills/export-open-issues $ npx degit huytieu/COG-second-brain/.agents/skills/generate-prd .claude/skills/generate-prd $ npx degit huytieu/COG-second-brain/.agents/skills/generate-release-notes .claude/skills/generate-release-notes $ npx degit huytieu/COG-second-brain/.agents/skills/harvest .claude/skills/harvest $ npx degit huytieu/COG-second-brain/.agents/skills/knowledge-consolidation .claude/skills/knowledge-consolidation $ npx degit huytieu/COG-second-brain/.agents/skills/loop-engineering .claude/skills/loop-engineering $ npx degit huytieu/COG-second-brain/.agents/skills/meeting-transcript .claude/skills/meeting-transcript $ npx degit huytieu/COG-second-brain/.agents/skills/memory-hygiene .claude/skills/memory-hygiene $ npx degit huytieu/COG-second-brain/.agents/skills/museum-art .claude/skills/museum-art $ npx degit huytieu/COG-second-brain/.agents/skills/no-ai-slop .claude/skills/no-ai-slop $ npx degit huytieu/COG-second-brain/.agents/skills/onboarding .claude/skills/onboarding $ npx degit huytieu/COG-second-brain/.agents/skills/product-ui-taste .claude/skills/product-ui-taste $ npx degit huytieu/COG-second-brain/.agents/skills/publish-to-confluence .claude/skills/publish-to-confluence $ npx degit huytieu/COG-second-brain/.agents/skills/release-video .claude/skills/release-video $ npx degit huytieu/COG-second-brain/.agents/skills/retro .claude/skills/retro $ npx degit huytieu/COG-second-brain/.agents/skills/review-cockpit .claude/skills/review-cockpit $ npx degit huytieu/COG-second-brain/.agents/skills/scout .claude/skills/scout $ npx degit huytieu/COG-second-brain/.agents/skills/slop-gate .claude/skills/slop-gate $ npx degit huytieu/COG-second-brain/.agents/skills/taste-skill .claude/skills/taste-skill $ npx degit huytieu/COG-second-brain/.agents/skills/team-brief .claude/skills/team-brief $ npx degit huytieu/COG-second-brain/.agents/skills/ultragoal .claude/skills/ultragoal $ npx degit huytieu/COG-second-brain/.agents/skills/update-cog .claude/skills/update-cog $ npx degit huytieu/COG-second-brain/.agents/skills/update-knowledge-base .claude/skills/update-knowledge-base $ npx degit huytieu/COG-second-brain/.agents/skills/url-dump .claude/skills/url-dump $ npx degit huytieu/COG-second-brain/.agents/skills/voice-baseline .claude/skills/voice-baseline $ npx degit huytieu/COG-second-brain/.agents/skills/weekly-checkin .claude/skills/weekly-checkin
$ curl -fsSL --create-dirs -o .claude/agents/brief-people-updater.md https://raw.githubusercontent.com/huytieu/COG-second-brain/main/.claude/agents/brief-people-updater.md $ curl -fsSL --create-dirs -o .claude/agents/fix-agent.md https://raw.githubusercontent.com/huytieu/COG-second-brain/main/.claude/agents/fix-agent.md $ curl -fsSL --create-dirs -o .claude/agents/harvest-curator.md https://raw.githubusercontent.com/huytieu/COG-second-brain/main/.claude/agents/harvest-curator.md $ curl -fsSL --create-dirs -o .claude/agents/integration-verifier.md https://raw.githubusercontent.com/huytieu/COG-second-brain/main/.claude/agents/integration-verifier.md $ curl -fsSL --create-dirs -o .claude/agents/task-verifier.md https://raw.githubusercontent.com/huytieu/COG-second-brain/main/.claude/agents/task-verifier.md $ curl -fsSL --create-dirs -o .claude/agents/worker-data-collector.md https://raw.githubusercontent.com/huytieu/COG-second-brain/main/.claude/agents/worker-data-collector.md $ curl -fsSL --create-dirs -o .claude/agents/worker-executor.md https://raw.githubusercontent.com/huytieu/COG-second-brain/main/.claude/agents/worker-executor.md $ curl -fsSL --create-dirs -o .claude/agents/worker-file-ops.md https://raw.githubusercontent.com/huytieu/COG-second-brain/main/.claude/agents/worker-file-ops.md $ curl -fsSL --create-dirs -o .claude/agents/worker-publisher.md https://raw.githubusercontent.com/huytieu/COG-second-brain/main/.claude/agents/worker-publisher.md $ curl -fsSL --create-dirs -o .claude/agents/worker-researcher.md https://raw.githubusercontent.com/huytieu/COG-second-brain/main/.claude/agents/worker-researcher.md
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 7,204 rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(rm -rf *)",
"Read(**/*.pem)",
"Bash(rm -rf /)",
"Bash(git push --force:*)",
"Bash(sudo *)",
"Read(.env)",
"Bash(rm -rf /*)",
"Read(./.env.*)",
"Read(~/.aws/**)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Read(**/*.key)",
"Read(**/.env.*)",
"Bash(sudo:*)",
"Bash(git reset --hard*)",
"Bash(git reset --hard:*)",
"Read(.env.*)"
],
"ask": [
"Bash(git push:*)",
"Bash(git push *)",
"Bash(git commit:*)",
"Bash(rm *)",
"Bash(rm:*)",
"Bash(git rebase *)",
"Bash(wget *)",
"Bash(npm publish:*)",
"Bash(git commit *)",
"Bash(gh pr merge *)"
]
}
} Skills (36)
auto-researchbraindumpclosed-loopcomprehensive-analysiscontent-factorycreate-user-storydaily-briefdaily-journaldata-formseditorial-illustrationsexport-open-issuesgenerate-prdgenerate-release-notesharvestknowledge-consolidationloop-engineeringmeeting-transcriptmemory-hygienemuseum-artno-ai-sloponboardingproduct-ui-tastepublish-to-confluencerelease-videoretroreview-cockpitscoutslop-gatetaste-skillteam-briefultragoalupdate-cogupdate-knowledge-baseurl-dumpvoice-baselineweekly-checkin
Subagents (10)
| brief-people-updater model: sonnet | Update people profiles in 05-knowledge/people/ with new information from brief data, meetings, or Slack |
| fix-agent model: sonnet | Targeted fixes after task-verifier returns FAIL:fixable. Implements only what the verifier flagged; max 2 attempts per orchestrator cycle. |
| harvest-curator model: sonnet | Draft session learnings from harvest staging into lizard-ready adoption notes. Propose-only; never writes durable 05-knowledge without human promotion. |
| integration-verifier model: sonnet | Read-only CP-4 gate. Cross-task wiring and global acceptance for multi-task specs. Cannot edit files. |
| task-verifier model: sonnet | Read-only verification gate. Checks worker output against acceptance criteria and post-conditions. Cannot edit files or mutate external state. |
| worker-data-collector model: sonnet | Collect data from GitHub, Slack, Jira, Linear, or file system. Structured extraction only — no synthesis. |
| worker-executor model: sonnet | Execute pre-approved mutations — Jira transitions, Linear updates, API calls, build commands. |
| worker-file-ops model: sonnet | Read, write, and organize vault files. Metadata updates, file moves, profile updates. |
| worker-publisher model: sonnet | Execute publishing operations — Slack, Confluence, Notion, webhooks. Receives final content and posts it. |
| worker-researcher model: sonnet | Web research agent. Searches, fetches URLs, extracts facts and evidence. No synthesis — just structured findings. |
Similar rigs
dwarvesf/dwarves-kit
The control plane for proof-based Claude Code SDLC, built for production teams. Agent loops that self-correct, not self-report.
Fort Knox 9.7k tok ·
atomicpapa/claude-obsidian-setup
A step-by-step guide to building an AI-powered second brain with Claude Code and Obsidian
Pragmatist 400 tok ·
darkroomengineering/cc-settings
Claude Code configuration — agents, skills, hooks & settings for Darkroom Engineering
Skill Collector 9.9k tok ·
yusufkaracaburun/ai-kit
The agile lifecycle as Claude Code + Cursor primitives. Stack-agnostic. One kit ships skills, subagents, slash commands, hooks, rules, and a Claude Code plugin manifest — for the two hosts ai-kit supports: Claude Code and Cursor.
Skill Collector 3.5k tok ·