agentleFS
TravisCaoTravisCao/cleanresearch/.claude/skills/experiment/SKILL.md

experiment

Skill1 starsChanged 3 months agoMITView on GitHub

Plan, run, freeze, and record a scientific experiment with a complete evidence trail. Use for any analysis, simulation, model run, data transformation, or other run that can affect a scientific conclusion.

29 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use cost, eval and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • maxvaegaSkill

    cost

    Estimate what Claude Cowork / Claude Code conversations would cost if run on the pay-per-token Claude API instead of the subscription. Reports token counts and cache-aware API cost in USD and EUR, for the current conversation or for the whole history (cost per conversation). Trigger when the user asks things like "how much would this conversation cost on the API", "API cost of my Cowork conversations", "token usage / token count", "cost per conversation", "/cost", or wants to compare subscription vs API pricing.

    Also in 3 projects that use this

  • qtzx06Skill

    eval

    Evaluate trained YOLO model accuracy with mAP metrics and per-class error analysis. Identifies failure cases for targeted improvement. Use after training.

    Also in 4 projects that use this

  • dirienSkill

    decide

    Three-agent adversarial debate protocol for strategic decisions. Two advocates steelman opposing positions while a moderator identifies risks and synthesizes a binding consensus. Use when choosing between alternatives, evaluating tradeoffs, or making high-stakes decisions.

    Also in 3 projects that use this

  • javatarzSkill

    wiki

    Access and manage the local GitHub Wiki clone. Use this skill when you need to read wiki content (architecture, design docs, guides), update wiki documentation, check wiki status, or push wiki changes. The wiki contains high-level system design and architectural documentation.

    Also in 3 projects that use this

  • HeshamFSSkill

    loop

    Run a command or check on a recurring interval

    Also in 3 projects that use this

  • VercelSkill

    workflow

    Creates durable, resumable workflows using Vercel's Workflow SDK. Use when building workflows that need to survive restarts, pause for external events, retry on failure, or coordinate multi-step operations over time. Triggers on mentions of "workflow", "durable functions", "resumable", "workflow sdk", "queue", "event", "push", "subscribe", or step-based orchestration.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 4 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 27 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Create the record before the run.

    line 8
  2. Distinctive:

    Read PROJECT.md, hypotheses.md, and memory/decisions.md.

    line 12
  3. Very specific:

    Create EXPERIMENT.md inside the folder.

    line 18
  4. Very specific:

    If the run errors, set status to failed.

    line 35
  5. Distinctive:

    Add one row to experiments/INDEX.md.

    line 45
  6. Distinctive:

    Check that every link resolves and that the frozen files match what ran.

    line 47

“Common” is measured: kept by at least 6 in 10 of the 27 edited versions of this file we could read. “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

Run a traceable experiment

Create the record before the run. Write the expected result before seeing the output.

1. Define the test

Read PROJECT.md, hypotheses.md, and memory/decisions.md. Confirm which hypothesis the run tests. Ask the researcher if the scientific aim is unclear.

Find the highest experiment number in experiments/INDEX.md. Create experiments/NNN-short-name/ with the next number.

2. Write the pre-run record

Create EXPERIMENT.md inside the folder. Copy the record fields from AGENTS.md. Before running, complete:

  • status: planned
  • hypothesis ID and link
  • date
  • expectation and the result that would count against it
  • exact command
  • input paths
  • configuration paths and key settings
  • planned output paths

The expectation must exist before any result does.

3. Run the test

Set status to running. Use code from src/. Run the exact recorded command. Keep raw data unchanged.

If the run errors, set status to failed. Record the error, where it happened, and any partial output. Add the failed experiment to the index. A failed run is normal scientific evidence. Do not hide it or reuse its number.

4. Freeze and interpret

Copy the exact executed code and configuration into the experiment folder. Record the frozen file paths. Do not treat the frozen copies as live code.

Set status to completed when the command finishes and the outputs are readable. Add the key numbers and output links under Result. Under Interpretation, state what the result supports, what it counts against, and what it cannot show. Write one concise Insight.

5. Connect the evidence

Add one row to experiments/INDEX.md. Add the insight to insights.md. Link the experiment from its hypothesis in hypotheses.md. Ask for explicit confirmation before changing the hypothesis text or state.

Check that every link resolves and that the frozen files match what ran. Create a Git checkpoint with this exact message pattern:

experiment NNN: <name>

Report the verdict, key numbers, evidence links, and one suggested next scientific step.

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.