agentleFS
cellrishi-codecellrishi-code/claude-skills-atlas/skills/research/evidence-checker/SKILL.md

evidence-checker

Skill5 starsChanged 43 days agoMITView on GitHub

Use when reviewing factual claims, research notes, technical statements, or literature summaries against supplied evidence.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    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:

    Never fabricate a source, citation, result, statistic, experiment, or quotation.

    line 28
  2. Distinctive:

    If the evidence is absent, say that it is absent.

    line 30
  3. Distinctive:

    Use this skill when the task involves evidence checker and the user needs a structured, repeatable approach.

    line 38
  4. Distinctive:

    Do not invent missing facts, data, citations, or project-specific behavior.

    line 64

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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

Evidence Checker

Goal

Prevent a draft from becoming stronger than its evidence.

For every important claim:

  1. Identify the claim.
  2. Locate supporting evidence in the supplied material.
  3. Classify it as:
    • directly supported
    • partially supported
    • contradicted
    • not established
  4. Explain the classification briefly.
  5. Identify missing context or alternative interpretations.
  6. Rewrite overly strong claims conservatively when requested.

Never fabricate a source, citation, result, statistic, experiment, or quotation.

If the evidence is absent, say that it is absent.

Purpose

Support users who need to reviewing factual claims, research notes, technical statements, or literature summaries against supplied evidence..

When to use

Use this skill when the task involves evidence checker and the user needs a structured, repeatable approach.

Instructions

Apply the guidance and checks above in sequence. Use the supplied context as the source of truth, make assumptions explicit, and verify important conclusions before presenting them.

Inputs

  • The user's task, question, or objective.
  • Relevant source material, data, constraints, and prior decisions.
  • The required audience, depth, and output format, when specified.

Outputs

  • A result tailored to the stated task and audience.
  • The key steps, evidence, or reasoning needed to make the result understandable.
  • Assumptions, checks, and unresolved limitations where relevant.

Example

Scenario: A user requests help with evidence checker and supplies the relevant context.

Expected result: Apply the guidance above, show the task-relevant result, and distinguish verified information from assumptions or open questions.

Limitations

Do not invent missing facts, data, citations, or project-specific behavior. State uncertainty where the supplied evidence is insufficient, and seek qualified human review when the task has material safety, legal, scientific, or operational consequences.

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.