| name | raise-vet |
| description | Use to vet a VC fund's legitimacy before spending real research on it — a cheap background check that answers "is this fund actually real and active, or a zombie?" Runs eight evidence-based checks and writes a stub fund dossier with a legitimacy score. Run before raise-match. Use when asked to "vet a fund", "check if a VC is legit", "is this fund still active", or "screen this investor". |
| allowed-tools | WebSearch WebFetch Read Write |
| metadata | {"track":"fundraising"} |
raise-vet — the cheap legitimacy gate
Most of a target list won't survive first contact — zombie funds with no dry
powder, firms that haven't led a round in two years, theses their portfolio
doesn't back up. raise-vet is the inexpensive pre-filter that catches them
before raise-match spends four research sub-agents on a dead end. It answers
"is this fund real and writing checks?" — not "does it fit us?" (that's
raise-match).
Invocation
raise-vet <fund name>
Steps
- Load the profile. Read
.spine/raise/profile.md for stage/sector
calibration (a fund can be alive but wrong for you — that's fine here, vet only
judges vitality). If it's missing, stop and tell the founder to run
raise-init. Derive the fund slug (see
../raise-init/references/namespace.md).
- Run the eight checks with a small number of targeted searches (this is the
cheap pass — don't fan out sub-agents):
- Vitality — date of last announced investment. >18 months silent = zombie risk.
- Health — active current fund with dry powder? Recent key-partner departures?
- Thesis authenticity — does the actual portfolio match the stated sector thesis?
- Lead capacity — evidence of leading rounds, not only co-investing.
- Stage honesty — stated stage vs. the real entry stage of recent deals.
- Check-size reality — stated range vs. observable deal sizes.
- Signal-to-check ratio — conference/content volume vs. actual new deals.
- Founder sentiment — public patterns: ghosting, slow decisions, reneging.
- Score and judge. Sum to a legitimacy score (0–100) and a verdict:
- legit (80–100) — active, real checks. Proceed to
raise-match.
- probable (60–79) — yellow flags; proceed if fit looks strong.
- watch (40–59) — material poser signals; de-prioritise unless a warm intro exists.
- poser (0–39) — don't spend cycles; move to the next fund.
- Write the stub dossier. Write
.spine/raise/funds/<slug>/dossier.md with
the structured handoff header (legitimacy_score, legitimacy_verdict, and a
fit_score/recommended_action left blank for raise-match) followed by a
## Legitimacy section: a one-line verdict, the eight checks as a table with
evidence and a 🟢/🟡/🔴 per check, and any open questions. Cite sources; mark
unknowns as Unknown, never invent dates or deal data. If a dossier already
exists, refresh only the header's legitimacy fields and the Legitimacy section
— never discard a raise-match analysis.
- Report. Tell the founder the score and verdict, and the next move: legit/
probable → "Run
raise-match <fund> for the fit analysis"; watch → "only if a
warm intro exists"; poser → "skip — run raise-funds for better-matched funds".
Spine I/O
- Reads:
.spine/raise/profile.md; .spine/raise/funds/<slug>/dossier.md (if present).
- Writes:
.spine/raise/funds/<slug>/dossier.md (header legitimacy fields + Legitimacy section).
Notes
- Vet is the cheap pass — a few searches, no sub-agent fan-out. The expensive
research is
raise-match's job, and it only runs on funds that clear this gate.
- Legitimacy ≠ fit. A fully legit fund can still be wrong for your stage; a
perfect-thesis fund can be a zombie. You need both reads.
raise-match is the sole assembler of dossier.md (see ADR 0017) — vet writes
only the stub it will build on.