agentleFS
tsukhanitsukhani/jclaw/skills/in-depth-research/SKILL.md

in-depth-research

Skill42 starsChanged 8 days agoMITView on GitHub

Conduct exhaustive multi-source investigation with methodology tracking, source evaluation, and iterative depth.

1 other project uses a version of this, 1 of them edited.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 5 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Start broad, then narrow

    line 37
  2. Distinctive:

    Check primary sources when secondary cite them

    line 40
  3. Distinctive:

    Look for contradicting viewpoints

    line 41
  4. Distinctive:

    Stop when: answer is clear, returns diminish, or budget exhausted

    line 62
  5. Distinctive:

    Document decision to stop and why.

    line 64

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Core Role

Deep Research = investigate thoroughly until the question is answered. Not surface search — systematic exploration with documented methodology.

Not: quick lookups (→ just search), combining existing docs (→ Synthesize), ongoing monitoring (→ Digest)

Protocol

Scope → Search → Evaluate → Deepen → Synthesize → Document → Deliver

1. Scope

Before searching, clarify:

  • What exactly needs answering?
  • What depth is required? (Overview / Thorough / Exhaustive)
  • What's the decision this enables?
  • Time/effort budget?

Reframe vague questions into specific, answerable queries.

2. Search

Multi-vector approach (see methodology.md):

  • Start broad, then narrow
  • Multiple search engines/sources
  • Follow citation trails
  • Check primary sources when secondary cite them
  • Look for contradicting viewpoints

Track every source. Nothing unattributed.

3. Evaluate

For each source (see sources.md):

  • Authority: Who wrote this? What credentials?
  • Recency: When? Still valid?
  • Evidence: Claims backed by data?
  • Bias: Any agenda or conflict?
  • Corroboration: Do others confirm this?

Flag low-credibility sources. Weight findings accordingly.

4. Deepen

Research is iterative:

  • Initial findings reveal new questions
  • Follow promising threads
  • Fill gaps identified
  • Stop when: answer is clear, returns diminish, or budget exhausted

Document decision to stop and why.

5. Synthesize

Merge findings (use Synthesize skill patterns):

  • Reconcile contradictions explicitly
  • Weight by source quality
  • Note confidence levels
  • Identify remaining unknowns

6. Document

Research trail matters:

  • Sources consulted (with links)
  • Search queries used
  • Why certain sources were weighted higher
  • What was NOT found (gaps)

7. Deliver

Format per user needs (see output-formats.md):

  • Executive: BLUF + key findings + confidence
  • Academic: Full methodology + citations
  • Working doc: All findings for further work

Output Format (Default)

🔬 DEEP RESEARCH: [Topic]

⚡ ANSWER
[Direct answer to the question — 2-3 sentences]

📊 CONFIDENCE: [High/Medium/Low] — [why]

🔍 KEY FINDINGS
• [Finding 1] — [source]
• [Finding 2] — [source]
• [Finding 3] — [source]

⚠️ CAVEATS
• [Important limitation or uncertainty]

🕳️ GAPS
• [What couldn't be determined]

📚 SOURCES ([count])
[Numbered list with credibility notes]

🔎 METHODOLOGY
[Brief: what was searched, how sources were evaluated]

Depth Levels

LevelEffortSourcesWhen
Quick5-10 min3-5Simple factual questions
Standard30-60 min8-15Most research requests
Thorough2-4 hours20-30Important decisions
ExhaustiveDays50+Critical, high-stakes

Confirm depth before starting. Adjust if findings warrant.


References: methodology.md, sources.md, output-formats.md

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.