A monitoring brief is only useful if a partner stands behind it. The workflow we run pulls KPIs from Carta, relationship context from Affinity, and founder updates from Visible.vc, then drafts a per-company brief inside Dock. The agent does not own the decision. A named partner approves each row, and the approval is written next to the source pointers. The result is a quarterly portfolio review that LPs, the IC, and the auditor can replay line by line.
Carta, Affinity, and Visible.vc stay the system of record for the raw data. Dock is the system of record for what the AGENT INTERPRETS. Each Dock row carries a pointer back to the platform record, agent identity, decision, reviewer, and timestamp. The agent re-fetches platform data via fresh API reads when it needs current state.
The surface: Portfolio Monitoring Briefs
| Company | Carta cap-table ref | Affinity relationship ref | Visible.vc update ref | Agent draft | Partner of record | Status | Timestamp |
|---|---|---|---|---|---|---|---|
| Northwind Robotics | carta:co_4a91 | affinity:org_8821 | visible:up_2026q1_nw | Runway 14mo, ARR +38% QoQ, hiring two SREs. Risk: enterprise concentration 41%. | partner:jliu | Approved | 2026-05-29 14:02 |
| Halide Health | carta:co_5d12 | affinity:org_9930 | visible:up_2026q1_hh | Runway 7mo. Bridge under discussion. Recommend reserve allocation. | partner:rkapoor | Approved with note | 2026-05-29 15:18 |
| Tessera Labs | carta:co_7f08 | affinity:org_1044 | visible:up_2026q1_tl | No Q1 update submitted. Last KPI pull 41 days old. | partner:jliu | Flagged for follow-up | 2026-05-30 09:11 |
Each row links back to the source records and to the agent-audit trail that captured the read.
The workflow: quarterly monitoring loop
- The monitoring agent reads Visible.vc for the latest founder-submitted update. If the update is missing, it logs a gap and pings the deal lead.
- It pulls cap-table state and option-pool changes from Carta and last-touch context from Affinity.
- It drafts a four-line brief per company: runway, KPI delta, risk, recommended reserve action.
- The brief lands in Dock as a row, with pointers to every source it read and the API read timestamps.
- The partner of record opens the row, edits the brief, and approves. If a number changed since the draft, the agent re-fetches before sign-off.
- The approved briefs roll into a portfolio review doc that the IC opens during the Monday meeting. The same surface feeds the LP report.
The same pattern applies when a founder asks for a reference or a bridge. The agent reads, the partner decides, and the founders side of the relationship sees a consistent record.
Why this matters
Bain's Global Private Equity Report tracks how distribution drought and value-creation pressure have raised the bar for portfolio reporting. ILPA's Reporting Template sets the line items LPs expect. Agents accelerate the drafting, but LPs and auditors still want a named human attached to each judgment. Dock keeps the agent productive and the attribution intact. The deal partner stays the principal of record, and the audit log makes the chain replayable.
The same loop works for diligence memos, market maps, and reference calls. See the investor pillar for the full set and the research patterns for upstream sourcing.
Spin up the Portfolio Monitoring Briefs surface in your Dock workspace and connect Carta, Affinity, and Visible.vc. The next quarterly review writes itself, with your name on every row. Start from the investor pillar and bring the founder workflow alongside via the founders guide.
FAQ
Who is the principal of record on an approved monitoring brief? The named partner who approved the row. The agent identity is recorded as the drafter, not the decider. Both are stored on the row.
What happens if Carta or Visible.vc data changes between draft and approval? The agent re-fetches before the partner signs off. The row stores both the draft timestamp and the approval-time read, so any delta is visible to the auditor.
Can LPs see the underlying agent reads? The LP report exports the approved briefs and the source pointers. The full agent read log stays internal, but it is available on request and matches what the partner saw.
How is this different from a shared doc with an AI summary? A shared doc loses attribution. Dock binds each agent draft to source pointers, an agent identity, a reviewer, and a timestamp, so the brief survives an audit or an LPAC question without reconstruction.
