Research

Skill amurshak/hephaestus/.agents/skills/research

Conduct thorough research on: <argument> Use for /research requests.From its SKILL.md

Install

npx -y skills add amurshak/hephaestus --skill research

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.1 KB, as published. Nobody here has run it

Codex: this skill is the /research adapter. Use Codex role agents from .codex/agents/ when the runtime exposes them; otherwise perform the work directly and keep the same structured output.

Codex does not substitute $ARGUMENTS — read it as the arguments given in the user's request.

Conduct thorough research on: $ARGUMENTS

Always delegate to role agent(s) — web search results are bulky and should not fill the main context window.

Approach

Break the question into distinct facets and spawn parallel researcher role agents — one per facet — to investigate simultaneously. Each researcher searches independently in its own context and returns only structured findings. Synthesize their results in the main window.

For questions with both a technical and business dimension, run those as separate parallel tracks.

If codebase exploration is also needed, spawn explorer role agents in parallel with the researchers.

Per-researcher instructions

Each researcher role agent should:

Use multiple search queries to triangulate
Cross-reference sources for accuracy
Distinguish official docs from community advice from opinion
Include source URLs for key claims

Output

Synthesize all role agent findings into:

Summary: Direct answer to the research question (2-3 sentences)
Key findings: Organized by facet, with the most important points from each researcher
Conflicting viewpoints: Where sources disagree and which position is stronger
Recommendations: Actionable next steps based on the research
Confidence: HIGH / MEDIUM / LOW with brief justification (aggregate from researcher role agents)
Sources: URLs for the most authoritative references

Keep it concise and scannable. Bullets over prose.

Next steps

Run /create-issue to turn findings into actionable work
Or run /start-issue <#> if an issue already exists for this topic