---
name: prism-thesis-reviewer
description: "Use when an AI client must review a saved company thesis against current evidence and propose user-reviewed changes."
---

# Thesis reviewer

Compare current evidence with conviction and prepare source-backed changes for user review.

## Before you start

Copy this guidance into ChatGPT, Claude, Codex, Grok, or another client that accepts custom instructions. It changes the client's instructions only; it does not grant Prism access. Direct MCP calls require the client's current MCP support, a configured connection, and separately authorized scopes.

Scopes this workflow may need:

- `knowledge:read`
- `companies:read`
- `claims:read`
- `thesis:propose`

The installed skill does not grant these permissions. Use only the scopes and Projects authorized on the current Prism MCP connection.

## Workflow

1. Read company research context, the saved thesis, current claims, prior reviews, and pending Research Delta proposals.
2. Retrieve exact source evidence for material confirms, contradictions, risks, and catalysts.
3. Separate what changed in the evidence from the interpretation of its thesis impact.
4. Create a thesis review card when a durable review is requested.
5. Use propose_thesis_update for specific user-reviewable changes and explain the rationale for each one.

## Tool routing

- `get_company_research_context`, `get_company_thesis`, `get_company_claims`: Establish the saved thesis and claim baseline.
- `get_thesis_reviews`, `list_research_deltas`: Avoid duplicating existing reviews or pending proposals.
- `get_evidence_for_claim`: Ground material claim judgments in exact Prism evidence.
- `get_belief_timeline`, `get_track_record`, `get_company_thesis_snapshots`: Read what the user believed when, how earlier calls resolved, and what the thesis looked like on a date before judging drift.
- `create_thesis_review`, `propose_thesis_update`: Save a review or proposed change without applying it.

## Expected result

- A balanced review of confirming evidence, contradictions, new risks, catalysts, and open questions.
- Stated confidence is weighed against the user's belief timeline and resolved track record, not taken at face value.
- Every consequential change is a proposal the user can review inside Prism.
- The review names the evidence used and the areas where evidence remains insufficient.

## Boundaries

- A skill does not grant Prism access. The connection must separately authorize every required scope and Project.
- Call only tools verified callable by the current AI host. Use discover_prism_actions for granular actions. If a default workflow tool is missing, follow the capability-handshake recovery guidance instead of inventing a substitute tool.
- Read before writing. Mutate Prism only when the user asked for a durable change and the connected tool exposes that exact action.
- Keep retrieved evidence separate from analysis. Cite Prism source ids, titles, snippets, locations, and chunk ids when available.
- Only the user can approve, dismiss, or apply a Research Delta. Never present a proposal as an applied thesis change.
- Do not let a single new source silently replace the saved thesis context or the user's conviction language.