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/auto-sync-sources.md
View on GitHub# Auto-sync sources
Company brains benefit from automated capture. Manual capture works for early-stage teams; auto-sync becomes necessary as call volume + team size scale.
**Start manual.** Automate only after 4-6 weeks of manual capture proves the team is actually using the brain.
## Common sources + setup patterns
### Fathom (call notes)
- **What**: AI-generated meeting summaries + full transcripts
- **Sync via**: Zapier / n8n webhook OR Fathom's built-in export → drop into `meetings/`
- **Naming**: `meetings/YYYY-MM-DD-<company-slug>-<slug>.md`
- **Metadata**: attendees list, duration, recording URL
- **Setup**: wire the webhook via `/toolify fathom` when ready
### Gong (sales calls)
- **What**: Enterprise sales call transcripts + coaching notes
- **Sync via**: Gong API (subscription tier dependent) → daily pull
- **Naming**: same as Fathom
- **Metadata**: deal ID, opportunity stage, coaching flags
- **Setup**: `/toolify gong` + schedule daily sync via `/loopify`
### Granola (personal + team meeting notes)
- **What**: Notes-first meeting app (transcript + your own live notes)
- **Sync via**: Granola export (manual or scheduled)
- **Naming**: `meetings/YYYY-MM-DD-<slug>.md`
- **Metadata**: your live-notes tags become wiki hints during compilation
### Slack (team discussions)
- **What**: Preserving discussion threads that resolve decisions or capture context
- **Sync via**: manual per-thread OR scheduled export of specific channels
- **Naming**: `raw/slack-<channel>-<date>.md`
- **Selection**: NOT every message — only threads that resolve a decision, capture a customer quote, or document a process
- **Recommendation**: `#decisions`, `#customer-signals`, `#post-mortems` channels get auto-synced; others stay manual
### Email (Front / Missive / Superhuman)
- **What**: Customer-facing threads worth preserving beyond the inbox
- **Sync via**: forward-to-address (dedicated inbox email address the vault auto-processes)
- **Naming**: append to `people/<person-slug>.md` OR `companies/<company-slug>.md` (thread goes with the person/org, not standalone)
- **Metadata**: thread ID, sender, subject line
- **Setup**: `/toolify <email-provider>` when the team is ready
### CRM (HubSpot / Attio / Pipedrive)
- **What**: Deal state, contact info, opportunity size
- **Sync via**: API pull (nightly or on-change webhook)
- **Naming**: updates `companies/<slug>.md` and `people/<slug>.md` in place (overwrites `status`, `last_touch`, `opportunity_size` fields)
- **Metadata**: opportunity stage, close date, ARR estimate
- **Setup**: `/toolify hubspot` + nightly sync via `/loopify`
### Support tickets (Intercom / Zendesk / Front)
- **What**: Customer questions + support patterns
- **Sync via**: API pull daily
- **Naming**: append repeat-question patterns to `recurring-questions/`; standalone tickets stay in the support tool
- **Selection filter**: only tickets tagged "escalated" or repeated 3+ times
## Cadence recommendations
Wire via `/loopify` with these defaults:
| Source | Cadence | Rationale |
|---|---|---|
| Fathom / Gong / Granola | Daily at 6am (after morning calls settle overnight, before team starts work) | Fresh capture, no overlap with active calls |
| CRM sync | Nightly at 1am | Off-peak, single-source-of-truth reconciliation |
| Slack export | Weekly (Sunday) | Weekly discussion digest fits weekly review cadence |
| Support ticket patterns | Weekly (Monday) | Fresh for support-team standups |
| Email forward-to-address | On-arrival (webhook) | Real-time — sales team wants context immediately |
## Idempotency
All auto-sync jobs must be idempotent. Company-brain has a lint check for this (repeated appends creating duplicate content). Follow the loopify guidance: dedupe by source ID / transcript hash / timestamp before appending.
## When NOT to auto-sync
- **Sensitivity mismatch**: don't auto-sync `confidential` content — manual capture only, with explicit access-list assignment
- **Team hasn't earned it**: if the team isn't capturing manually, auto-sync creates a data-firehose nobody uses. Prove capture demand first.
- **Rate-limited APIs**: some sources (older CRM APIs, Slack free-tier export) rate-limit hard. Manual > broken auto-sync.
- **Compliance-restricted**: for regulated industries (healthcare, legal, finance), auto-sync of client data may violate compliance. Consult legal first.
## Recipe library (populate as team wires them)
- `references/recipes/fathom-webhook-nextjs.md` — Fathom webhook receiver in Next.js
- `references/recipes/gong-daily-sync.md` — Gong API pull loop
- `references/recipes/attio-crm-sync.md` — Attio nightly sync
- `references/recipes/slack-weekly-export.md` — Slack export → structured raw
Each recipe pairs `/toolify <source>` (initial wiring) + `/loopify` (scheduled sync).
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 | Verified | 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 |