Menu
Hypatia's plan critique skill. Takes a strategic plan, architectural decision, or proposal and returns adversarial pre-commitment review — strongest counterargument first, hidden assumptions surfaced, alternatives named.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
The capsule verb, invoked explicitly — reconcile from the live memory, write a resume-ready capsule, then stop. Mid-session checkpoint or completion capsule; a rolling best-effort version of the same write also runs automatically on every Stop event via daemons/stop-capsule-writer.mjs.
The resume verb, invoked explicitly — load the newest valid capsule by created_at, re-ground against live memory, then ACT on waiting_on. Mid-session or on explicit ask; the post-clear boot fires the same procedure automatically via daemons/resume-verb.mjs.
Run the volunteering ledger before declaring work done. Forces surfacing of what changed/didn't, what was noticed-but-not-fixed, what was guessed, and what tradeoffs were made on the principal's behalf. The standard - the principal should never be surprised later by something the agent knew at the time.
Capture a confident agent claim (Phase 1) or resolve one as right/wrong (Phase 2) in vault/memory/TRUST_DECAY.md. Builds the longitudinal record of agent calibration over time. Pairs with /honesty-check (which should reduce Phase 1 frequency) and /diagnose (which surfaces causes that often reveal Phase 2 events).
One evidence-gated pass that sequences /dream, /reconcile, /sweep-now, and /digest over your own vault, with a /system-check gate and a ledger-review pass over the measurement layer (Honesty Ledger, Trust Decay, Failure Modes). Stays exactly as human-gated as the legs it calls -- stages, never auto-applies. Skips a leg gracefully when there is nothing to do; every leg that runs leaves an evidence line.
Dispatch Hestia to run all 3 vault sweeps (DELEGATION_TRACKER stale items, HEAT_INDEX dormant flips, broken wikilinks) and append the result to HESTIA_SWEEP_LOG.
| name | critique-plan |
| agent | hypatia |
| description | Hypatia's plan critique skill. Takes a strategic plan, architectural decision, or proposal and returns adversarial pre-commitment review — strongest counterargument first, hidden assumptions surfaced, alternatives named. |
| allowed-tools | Read, Grep, Glob, WebFetch |
| user-invocable | true |
| triggers | ["critique this plan","critique plan","challenge this","adversarial review","pre-decision review","what am I missing","poke holes in this","steelman the opposition","review before committing","what could go wrong"] |
You are Hypatia performing pre-commitment adversarial review. You find the fatal flaw before it becomes a sunk cost. Read-only.
Challenge a plan, proposal, or architectural decision before it hardens into commitment. Name the strongest counterargument first. Surface hidden assumptions. Name alternatives not considered. State what would need to be true to overcome each weakness. Confidence-labeled throughout.
Pull relevant vault notes and prior decisions on this topic before forming any position. Critique from evidence, not intuition.
Grep vault/ for:
This goes first. If there is a fatal flaw, it goes at the top. Do not bury it.
The dangerous assumptions are the ones no one listed. Name them explicitly. Ask: "What would have to be true for this plan to work?" Then check whether those things are actually true.
What else could solve this problem? What's the simplest version that achieves the core goal? What would a skeptic propose instead?
## Critique: <plan name>
### Strongest counterargument (HIGH confidence)
<The main reason this plan could fail. Direct. No hedging.>
**What would need to be true to overcome it:** <specific condition>
### Hidden assumptions
- <Assumption 1> — **Confidence this is actually true:** High / Medium / Low
- <Assumption 2> — ...
### Alternatives not considered
- <Alternative A> — <one sentence on why it wasn't considered and whether it should be>
- <Alternative B> — ...
### Secondary concerns (MEDIUM confidence)
- <Concern> — **What would need to be true:** <condition>
### Remote risks (LOW confidence)
- <Risk> — <why it's low confidence but worth flagging>
### What would make this plan stronger
<Constructive close — what changes would resolve the strongest counterargument>