agentleFS
Sign inStart free
equinorequinor/fusion-skills/.github/copilot-instructions.md

fusion-skills

Copilot instructions2 starsChanged 2 months agoMITView on GitHub

Repository-wide guidance for Copilot in this repository. Write skill prose for minimal token cost. Every token in SKILL.md is loaded into context on activation. Copilot is a repository contributor and must follow the same standards as human contributors, including CONTRIBUTING.md and maintainers guidance in contribute/ for all relevant changes. - Keep changes minimal and scoped to the request. - Never request, expose, or store secrets/credentials. - Do not run destructive commands without explicit user confirmation. - Do not invent validation…

The file

We split this copilot-instructions.md into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    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
  • could be automatic: a setting could enforce it
  1. Distinctive:

    Keep this file under 100 lines and applicable to all repository contexts.

    line 7
  2. Very specific:

    Put detailed, domain-specific rules in .github/instructions/*.instructions.md.

    line 8
  3. Very specific:

    Write skill prose for minimal token cost.

    line 32
  4. Distinctive:

    Use fragments: "Run tests before commit" not "You should always run tests before committing"

    line 35
  5. Distinctive:

    Keep changes minimal and scoped to the request.

    line 52
  6. Distinctive:

    Never request, expose, or store secrets/credentials.

    line 53
  7. Distinctive:

    Do not run destructive commands without explicit user confirmation.

    line 54
  8. Distinctive:

    Do not invent validation outcomes.

    line 55

≈ “Common” is an estimate: at least 1 in 5 of the 3,779 Copilot instructions 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
copilot-instructions.md

Copilot Instructions

Repository-wide guidance for Copilot in this repository.

Scope

  • Keep this file under 100 lines and applicable to all repository contexts.
  • Put detailed, domain-specific rules in .github/instructions/*.instructions.md.

Path-specific instructions

PathInstructions file
skills/**.github/instructions/skills-authoring.instructions.md
skills/**/scripts/**.github/instructions/skills-scripts-safety.instructions.md
scripts/**.github/instructions/scripts-code-rules.instructions.md
Pull request prep/updates.github/instructions/pr-workflow.instructions.md

Skill discovery defaults

  • For local skill discovery in this repository, search skills/ first.
  • Include skills/ and skills/.experimental/ in local .agents/skills lookup paths for this workspace.

SKILL.md size limits

  • SKILL.md files must stay under 500 lines (CI error threshold).
  • SKILL.md files above 300 lines trigger a CI warning — consider moving content to references/.
  • See contribute/02-skill-structure-and-content.md for the cross-platform size rationale and progressive disclosure guidance.

SKILL.md prose style

Write skill prose for minimal token cost. Every token in SKILL.md is loaded into context on activation.

  • Drop articles (a/an/the), filler (just/really/basically/simply), and hedging
  • Use fragments: "Run tests before commit" not "You should always run tests before committing"
  • Short synonyms: "fix" not "implement a solution for", "use" not "utilize"
  • State actions directly — drop "you should", "remember to", "make sure to"
  • Merge redundant bullets; keep one example when multiple show the same pattern
  • In SKILL.md files, never modify code blocks, inline code, paths, URLs, or technical terms
  • To compress any skills/**/*.md file, use the caveman-compress skill (.agents/skills/caveman-compress): /caveman:compress skills/<name>/SKILL.md

Maintainer docs

  • For maintainer workflow and policy context, use contribute/README.md and the topic files in contribute/.

Contributor parity

Copilot is a repository contributor and must follow the same standards as human contributors, including CONTRIBUTING.md and maintainers guidance in contribute/ for all relevant changes.

Global guardrails

  • Keep changes minimal and scoped to the request.
  • Never request, expose, or store secrets/credentials.
  • Do not run destructive commands without explicit user confirmation.
  • Do not invent validation outcomes; run commands and report actual results.
  • For skill create/update PRs, never manually edit generated release artifacts:
    • root README.md
    • root CHANGELOG.md
    • any skills/**/CHANGELOG.md These are maintained by release automation.

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