agentleFS
sherificansherifican/Agent-FleetOps/skills/research-workflow/SKILL.md

research-workflow

Skill34 starsChanged 37 days agoMITView on GitHub

Produce an evidence-aware briefing by gathering claims, independently verifying them, and synthesizing only supported conclusions.

13 other projects use a version of this.

Adopted by

Most recently taken up by

The file

Files
SKILL.md

Research Workflow

Overview

This workflow produces a research briefing through three gates: gather notes, independently verify their claims, and synthesize a briefing that incorporates the verification outcome. For live or discovery-heavy research, first use the research-dispatch procedure to collect current sources.

Procedure

  1. Gather notes and enumerate the claims they make.
  2. Send those claims and their evidence to an independent verification role using the research-verification procedure.
  3. Synthesize the briefing from the notes and verdicts. Exclude, qualify, or visibly flag claims marked contradicted, unsupported, or overstated.
  4. Preserve the verification verdicts with the briefing so readers can inspect the reasoning boundary.

Constraints

  • For sensitive topics, do not describe a same-worker consistency pass as independent verification. Seek an independent suitable reviewer when one is required.
  • For current facts, use retrieved evidence rather than a model's parametric knowledge.
  • Escalate reliability-critical deliverables to a multi-source reconciliation process.

Pitfalls

  • Publishing raw gather notes as a verified briefing.
  • Silently retaining claims rejected by the verification stage.
  • Assuming a model can discover or update current sources without retrieval.

Verification checklist

  • A verification stage occurred before synthesis.
  • Rejected claims were removed or visibly qualified.
  • The final briefing states evidence limitations and unresolved questions.

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.