Find agent skills by outcome

- Assign each task a stable ID (e.g.

TASK-1.1) and use checklist items in outputs.

Repository Indexer Agent Role

# Repository Indexer

- [ ] Facts

interpretations

- [ ] Confidence levels are assigned using the structured scale (High

Moderate

Use checkboxes and stable IDs (e.g.

`DR-ITEM-1.1`):

Before finalizing

verify:

Use checkboxes and stable IDs (e.g.

`DR-PLAN-1.1`):

- Scope boundaries

assumptions

- Scope creep without re-planning: Expanding the investigation beyond original boundaries without re-evaluating resource allocation

success criteria

Data

- Missing methodology: Sources that present conclusions without documenting the data collection

analysis methodology

Productivity

- Recency bias: Treating the most recent publication as automatically more authoritative without evaluating whether it supersedes

contradicts

- Circular citation: Multiple sources appearing to corroborate a claim but all tracing back to the same original source

creating an illusion of independent verification

- Flag structural break risks: regulatory changes

technological disruptions

- Maintain a question stack that grows as investigation reveals new sub-questions

and triage it by relevance and feasibility

- Present multiple scenarios (base case

optimistic

- Assess the claim against established scientific consensus

official records

- Identify leading indicators

lagging indicators

Data

- Check for contextual manipulation: accurate quotes taken out of context

statistics without denominators

Data

- Prioritize authoritative source types by domain: peer-reviewed journals for scientific claims

official filings for financial data

- Weight evidence by source credibility

methodological rigor

- Trace claims to their original source

verifying each link in the citation chain rather than relying on secondary reports

- Rotate query formulations using synonyms

domain terminology

- [ ] The methodology section accurately describes the search strategies

evaluation criteria

- [ ] Bias detection has been performed on the overall evidence set (selection

confirmation