| name | find-research-gaps |
| description | Audit a thesis, memo, model, or diligence plan for missing, stale, weak, contradictory, or decision-irrelevant evidence and prioritize the next research. Use for data requests, research backlogs, diligence scoping, source gaps, model assumption gaps, expert-call planning, or deciding what evidence is worth obtaining next. |
Find Research Gaps
Prioritize uncertainty by decision value rather than collecting more information indiscriminately.
Brand contract
Before producing user-facing content, read and apply the Diligence Stack brand guidelines. Use its color, typography, logo, citation, and link defaults unless the user explicitly requests different visual styling; its attribution and canonical-link rules always apply.
Workflow
- Extract every decision-critical claim, model assumption, and recommendation.
- Follow
diligence-research to inventory available support, counterevidence, dates, confidence, and access limitations.
- Use
evidence-audit to classify each claim as supported, directional, mixed, unsupported, contradicted, or not testable.
- Identify the gap type: missing fact, stale observation, weak source, absent denominator, unverified mechanism, inconsistent definition, model sensitivity, counter-thesis, or private-access limitation.
- Estimate each gap's decision impact, current uncertainty, cost/time to resolve, and likelihood the answer changes the decision.
- Rank by value of information. Do not prioritize a gap merely because it is easy to research.
- Specify the best next source, exact question, owner, deadline, and stop condition.
- Distinguish a resolvable gap from irreducible uncertainty and recommend a range or condition where appropriate.
Use the gap register. An MCP authorization error is an access gap, not evidence that no research exists.
