Operator insight (2026-07-11): there are synergies between the second brain and the agents that will handle client projects. This anchors it in the PRD:
Knowledge layer (agent-architecture.md): the vault's client dossiers (01-Projects/Clients/<client>/) are the junction of the three agent families — back-office fleet (deposits billing state/contract facts; retrieves dunning tone), sales chain (prospection → proposal → contract), and future delivery agents (deposit decisions, meeting notes, new-business sightings — the KM 4 % settlement clause makes delivery observations a billing input; retrieve contract scope + billing state). Deposits = agent-owned files with ai_generated frontmatter, never edits of human notes; vault routing doctrine (Ornith/local first) for confidential client content.
Backlog (task-inventory.md): updated with the 2026-07-11 CRM audit facts (Dolibarr propal/contrat/projet modules enabled but empty; the KM retainer terms — $5,500/mo = $2,500 net-30 + $3,000 deferred +3 months — exist only as trunk WIP JSON, and that deferred schedule is the source of truth T06 dunning needs) and the third family as an explicit backlog line (local capture lane → vault ingest as an sb.py job; REX → Mode B evidence → CMS case studies).
Meeting capture decision (agent-architecture.mdD9 + agent-catalog.md T17): Hyprnote free tier (rebranded "Char") — speaker diarization and Google Calendar sync are both on the free plan, satisfying the two operator criteria at once; manual trigger only (consent stance — auto-record is a Pro feature we don't want anyway). Meetily (MIT, diarization in the community core) stays as OSS fallback, with sb.py ICS-matching to compensate its missing calendar sync. Gate before the lane is trusted: diarization quality judged on a real bilingual call (EN client / FR partner) — steps carried by erp#49.
Advances PRD:vibe/PRD/ai-back-office. QA: 356 links/anchors re-checked at authoring, zero broken; D9 additions anchor-checked.
## What
Operator insight (2026-07-11): there are synergies between the second brain and the agents that will handle **client projects**. This anchors it in the PRD:
- **Knowledge layer** (`agent-architecture.md`): the vault's client dossiers (`01-Projects/Clients/<client>/`) are the **junction of the three agent families** — back-office fleet (deposits billing state/contract facts; retrieves dunning tone), sales chain (prospection → proposal → contract), and future **delivery agents** (deposit decisions, meeting notes, *new-business sightings* — the KM 4 % settlement clause makes delivery observations a billing input; retrieve contract scope + billing state). Deposits = agent-owned files with `ai_generated` frontmatter, never edits of human notes; vault routing doctrine (Ornith/local first) for confidential client content.
- **Backlog** (`task-inventory.md`): updated with the 2026-07-11 CRM audit facts (Dolibarr propal/contrat/projet modules enabled but **empty**; the KM retainer terms — $5,500/mo = $2,500 net-30 + $3,000 deferred +3 months — exist only as trunk WIP JSON, and that deferred schedule is the source of truth T06 dunning needs) and the **third family** as an explicit backlog line (local capture lane → vault ingest as an `sb.py` job; REX → Mode B evidence → CMS case studies).
- **Meeting capture decision** (`agent-architecture.md` **D9** + `agent-catalog.md` T17): **Hyprnote free tier** (rebranded "Char") — speaker **diarization** and **Google Calendar sync** are both on the free plan, satisfying the two operator criteria at once; manual trigger only (consent stance — auto-record is a Pro feature we don't want anyway). **Meetily** (MIT, diarization in the community core) stays as OSS fallback, with `sb.py` ICS-matching to compensate its missing calendar sync. Gate before the lane is trusted: diarization quality judged on a real bilingual call (EN client / FR partner) — steps carried by erp#49.
**Advances PRD:** [vibe/PRD/ai-back-office](https://gitea.arcodange.lab/arcodange-org/factory/src/branch/main/vibe/PRD/ai-back-office/README.md). QA: 356 links/anchors re-checked at authoring, zero broken; D9 additions anchor-checked.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Operator insight 2026-07-11: the second brain's client dossiers
(01-Projects/Clients/<client>/) are where the back-office fleet meets
the sales chain and the future delivery agents working on client
projects. Knowledge-layer bullet: agent-owned deposit files (never
edits of human notes), each family feeds and reads the dossier
(billing state + contract facts in; dunning tone, contract scope,
new-business sightings out — the KM 4% settlement clause makes
delivery observations a billing input); vault routing doctrine
(local-first) for confidential client content. Backlog updated with
the 2026-07-11 CRM audit facts (propal/contrat/projet modules empty,
KM retainer terms only in WIP JSON — deferred schedule = T06's source
of truth) and the third family (delivery agents; Granola→vault ingest
as an sb.py job; REX → Mode B evidence → CMS).
Co-Authored-By: Claude Fable 5 <[email protected]>
Operator has no Granola account (proprietary, paid, cloud
transcription — misfit with the vault doctrine). The delivery-agents
backlog line now specifies the local transcription lane: Whisper-class
model on the M4 + Ornith summary, as an sb.py job.
Co-Authored-By: Claude Fable 5 <[email protected]>
Diarization and Google Calendar sync are both on Hyprnote's free plan,
which satisfies the two operator criteria at once; Meetily (MIT,
diarization in the community core) stays as OSS fallback with sb.py
ICS-matching to compensate its missing calendar sync. Gate: quality
judged on a real bilingual call before the lane is trusted (erp#49).
Co-Authored-By: Claude Fable 5 <[email protected]>
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
What
Operator insight (2026-07-11): there are synergies between the second brain and the agents that will handle client projects. This anchors it in the PRD:
agent-architecture.md): the vault's client dossiers (01-Projects/Clients/<client>/) are the junction of the three agent families — back-office fleet (deposits billing state/contract facts; retrieves dunning tone), sales chain (prospection → proposal → contract), and future delivery agents (deposit decisions, meeting notes, new-business sightings — the KM 4 % settlement clause makes delivery observations a billing input; retrieve contract scope + billing state). Deposits = agent-owned files withai_generatedfrontmatter, never edits of human notes; vault routing doctrine (Ornith/local first) for confidential client content.task-inventory.md): updated with the 2026-07-11 CRM audit facts (Dolibarr propal/contrat/projet modules enabled but empty; the KM retainer terms — $5,500/mo = $2,500 net-30 + $3,000 deferred +3 months — exist only as trunk WIP JSON, and that deferred schedule is the source of truth T06 dunning needs) and the third family as an explicit backlog line (local capture lane → vault ingest as ansb.pyjob; REX → Mode B evidence → CMS case studies).agent-architecture.mdD9 +agent-catalog.mdT17): Hyprnote free tier (rebranded "Char") — speaker diarization and Google Calendar sync are both on the free plan, satisfying the two operator criteria at once; manual trigger only (consent stance — auto-record is a Pro feature we don't want anyway). Meetily (MIT, diarization in the community core) stays as OSS fallback, withsb.pyICS-matching to compensate its missing calendar sync. Gate before the lane is trusted: diarization quality judged on a real bilingual call (EN client / FR partner) — steps carried by erp#49.Advances PRD: vibe/PRD/ai-back-office. QA: 356 links/anchors re-checked at authoring, zero broken; D9 additions anchor-checked.
🤖 Generated with Claude Code