coreyhaines31/makerskills / skills/company-brain
company-brain
VerifiedYour team's shared, AI-ready knowledge base — people, companies, meetings, SOPs, and decisions structured so an agent can answer on your team's behalf. Team-scope sibling to second-brain — personal notes go there, not here. Modes — capture, compile (wiki pages + INDEX.md), query (trust-weighted, saved to outputs/), review (verify / deprecate / supersede stale captures), lint, connect, search. Structured raw dirs (people/, companies/, meetings/, sops/, decisions/, customer-language/, sales-objections/). Every capture stamps author, timestamp, and trust status. Optional sync from call transcripts, Slack/email exports, CRM. Vault at ${COMPANY_BRAIN_VAULT:-$HOME/Documents/CompanyBrain}/. Triggers on "/company-brain," "/cb," "capture this into the team brain," "log this meeting," "save this SOP," "compile the company wiki," "query the team brain," "what does the team know about X," "review the company brain," "lint the company brain," "who's the internal expert on X."
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 19.3 KB |
| references/auto-sync-sources.md | markdown | 4.7 KB |
| references/rollout.md | markdown | 1.9 KB |
| references/schema.md | markdown | 8.2 KB |
| references/vault-config.md | markdown | 4.0 KB |
references/vault-config.md
View on GitHub# Vault config
## Location
Company brain lives at `${COMPANY_BRAIN_VAULT:-$HOME/Documents/CompanyBrain}/`.
Override in `~/.zshenv`:
```bash
export COMPANY_BRAIN_VAULT="$HOME/code/acme-company-brain" # per-team path
```
For multi-team users (e.g., a consultant maintaining vaults for multiple clients):
```bash
export COMPANY_BRAIN_VAULT="$HOME/code/current-client-brain"
# Switch by aliasing:
alias cb-acme="export COMPANY_BRAIN_VAULT=$HOME/code/acme-brain"
alias cb-beta="export COMPANY_BRAIN_VAULT=$HOME/code/beta-brain"
```
The skill respects whatever `$COMPANY_BRAIN_VAULT` points to at invocation time.
## Storage backends
Company brains live in one of three places, in order of team-scale readiness:
| Backend | When to use | Trade-offs |
|---|---|---|
| **Obsidian vault (local, git-synced)** | 1-10 person teams | Simplest, portable, git handles history + attribution. Requires everyone to install Obsidian. |
| **Shared git repo** (no Obsidian) | 5-50 person teams, engineering-heavy | Everyone reads markdown in their editor of choice. Git is the sync layer. Requires markdown literacy. |
| **Notion / Coda backed by markdown export** | 20+ person teams, non-technical majority | Team edits in Notion; nightly export → git-backed markdown for company-brain. Extra sync layer, but non-engineers can contribute. |
Company-brain skill treats all three the same — reads/writes markdown. The backend choice is a team decision, not a skill decision.
## Initial setup for a new team
```bash
# 1. Pick a path
export COMPANY_BRAIN_VAULT="$HOME/code/mycompany-brain"
# 2. Create the vault
mkdir -p "$COMPANY_BRAIN_VAULT"
cd "$COMPANY_BRAIN_VAULT"
# 3. Init git (if not already)
git init
git remote add origin git@github.com:mycompany/company-brain.git
# 4. Create the base dirs, plus ONLY the pilot workflow's dirs (see "Start narrow" below)
mkdir -p raw wiki outputs Projects Team Templates Drafts
mkdir -p meetings sales-objections customer-language # e.g. a Sales pilot
touch meetings/.gitkeep sales-objections/.gitkeep customer-language/.gitkeep # git doesn't track empty dirs; teammates' clones need them
# 5. Seed the schema doc from this skill's references/schema.md (path depends on where your agent installed the skill)
# AGENTS.md is read by Codex, Cursor, and most hosts; the symlink covers Claude Code
cp <skill-dir>/references/schema.md AGENTS.md
ln -s AGENTS.md CLAUDE.md
# 6. Seed INDEX.md
echo "# Wiki Index\n\n## Sales\n\n## Customers\n\n## Ops\n\n## Product\n\n## Team & People\n\n## Decisions\n\n## Playbooks" > wiki/INDEX.md
# 7. Seed .gitignore
cat > .gitignore <<EOF
.DS_Store
.obsidian/workspace*
.obsidian/cache
.trash/
Drafts/*.private.md
EOF
# 8. First commit
git add . && git commit -m "Seed company brain schema"
git push -u origin main
```
### Start narrow
Don't seed all eight structured dirs on day one. Pick the **one workflow closest to revenue** and seed only its dirs:
| Pilot workflow | Dirs to seed |
|---|---|
| Sales | `meetings/` + `sales-objections/` + `customer-language/` |
| Content / marketing | `customer-language/` + `recurring-questions/` + `sops/` |
| Ops / delivery | `sops/` + `decisions/` + `meetings/` |
| Reporting | `decisions/` + `companies/` + `meetings/` |
Other dirs get created by the first capture that needs one. Fewer dirs means a smaller review queue, so the review habit sticks before volume arrives. The brain earns the right to touch more. See [rollout.md](rollout.md) for widening it beyond the pilot group.
## Sensitivity access lists (optional)
If using sensitivity tagging, maintain two access lists in `Team/`:
```
Team/leadership.md # emails/handles that can read `leadership` content
Team/confidential-access.md # per-topic access lists for confidential wiki pages
```
Company-brain skill reads these when respecting sensitivity in `query` mode.
## Backup
Same as any git repo. Push to a hosted remote (GitHub / GitLab / Bitbucket). Recover by cloning.
For sensitive content in the vault, treat it like any git repo with secrets — encrypted at rest via `git-crypt` if the remote isn't fully trusted.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Unverified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |