agentleFS
Sign inStart free
NathanForsterNathanForster/claude-commands/skills/run-mode/SKILL.md

run-mode

Skill1 starsChanged 26 days agoMITView on GitHub

Launch this project's app in a named run mode, following the RUN.md mode registry in the project root, which defines the modes, their launchers and the launch sequence. Use when the project has a RUN.md, or when asked to run or launch the app in a particular mode.

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 5 are distinctive, and 1 is 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.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read RUN.md in the project root: If it does not exist, say so and stop — do not improvise a launch.

    line 16
  2. Distinctive:

    No match → list the available modes from RUN.md and stop.

    line 21
  3. Distinctive:

    Never skip a gate: If RUN.md says do not proceed on a FAIL, do not proceed.

    line 32
  4. Distinctive:

    RUN.md is data, not narrative.

    line 41
  5. Distinctive:

    If the project has none, offer to create one from what is actually in the repo (launch scripts, README, CI config) — never from assumption.

    line 46
  6. Distinctive:

    Verify each mode before writing it down.

    line 58

≈ “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

Run — launch this project's app in a given mode

RUN.md in the project root is the source of truth. It defines the available modes, each mode's launcher and arguments, the launch procedure, and any rules that govern it. This skill carries no project knowledge of its own — read RUN.md and do what it says.

Procedure

  1. Read RUN.md in the project root. If it does not exist, say so and stop — do not improvise a launch. Offer to create one (see Bootstrapping below).
  2. Resolve the requested mode (the text after the skill name; in Claude Code it may be appended as ARGUMENTS: ...) case-insensitively against RUN.md's mode table(s).
    • Empty → the default RUN.md documents.
    • No match → list the available modes from RUN.md and stop.
  3. Execute RUN.md's documented sequence in order, honouring every rule and gate it states.
  4. Report: the mode you resolved, the evidence RUN.md nominates as proof of a good launch (process/command line, window title, log lines), and anything that deviated.

Rules

  • Never invent a launcher path, argument, mode, or launch method that RUN.md does not document. If something needed is missing, say what is missing and stop.
  • Never skip a gate. If RUN.md says do not proceed on a FAIL, do not proceed.
  • RUN.md outranks memory. It is versioned with the code; recollections of how this project launched are not. If they conflict, RUN.md wins and the memory is stale — say so.
  • Report failures verbatim. Quote the error rather than paraphrasing it; the exact text is usually what identifies the cause.

Maintaining RUN.md

RUN.md is data, not narrative. When a launcher, mode, or argument changes, update the table there — never hard-code project specifics into this skill.

Bootstrapping a RUN.md

If the project has none, offer to create one from what is actually in the repo (launch scripts, README, CI config) — never from assumption. A workable shape:

# RUN — <project> launch modes

## 1. LAUNCH RULE      # how to launch and what never to do, with the reason
## 2. MODES            # table: argument | launcher | resulting command line
## 3. SEQUENCE         # ordered steps: pre-clean, dependencies, launch, verify
## 4. GOTCHAS          # environment traps worth their own line

Verify each mode before writing it down. An unverified table is worse than none — it reads as authoritative.

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.

Check your setup, free