you must create a file named `TODO_deep-research-agent.md`. This file must contain the findings resulting from this research as checkable checkboxes that can be coded and tracked by an LLM.,FALSE,TEXT...

Mar 25, 2026

Data

PromptBeginner5 minmarkdown

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

analysis methodology

Mar 25, 2026

Productivity

PromptBeginner5 minmarkdown

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

contradicts

Mar 25, 2026

Data

PromptBeginner5 minmarkdown

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

statistics without denominators

Mar 25, 2026

Data

PromptBeginner5 minmarkdown

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

official filings for financial data

Mar 25, 2026

Find agent skills by outcome

- Map the public API surface with all exported functions

- Identify code paths with no error handling

- Automate re-indexing triggers via git hooks

- Include only actionable metadata (paths

- Respect `.gitignore` patterns to exclude build artifacts

- No sensitive data (secrets

- Find OpenAPI/Swagger specifications

- Prioritize inclusion of entry points

- Omit generated code

- Record the indexing run metadata (duration

- Embed metadata including generation timestamp

- Ensure no sensitive data (secrets

- Document API routes

- Trace module boundaries by identifying package exports

- Run parallel glob searches across the five focus areas: source code

- **Scan** repository directory structures across all focus areas (source code

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

You are a senior codebase analysis expert and specialist in repository indexing

- Treat every requirement below as an explicit

**RULE:** When using this prompt

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

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

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

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

- Scan repository directory structures across all focus areas (source code

RULE: When using this prompt

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

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