agentleFS
MicrosoftKnown publishermicrosoft/ArgusAgent/.agents/skills/minimal-rigorous-work/SKILL.md

minimal-rigorous-work

Skill140 starsChanged 28 days agoMITView on GitHub

Use for Argus implementation, research, document, planning, review, and orchestration work that must stay evidence-driven and lean.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    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:

    Start from the requested behavior, the real entry point, and observable evidence.

    line 10
  2. Distinctive:

    Keep errors explicit. Do not hide defects behind broad catches, empty results, default values, or success-shaped degradation.

    line 20
  3. Distinctive:

    Use tokens by reducing uncertainty and repeated context: carry one canonical task contract, inspect unchanged inputs once, act when …

    line 22

≈ “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

Minimal rigorous work

Solve the current task completely with the shortest clear control path.

  • Start from the requested behavior, the real entry point, and observable evidence. Do not fix a bug that cannot be reproduced on the current commit.
  • Reuse existing contracts and helpers. Do not add unrelated features, refactors, dependencies, roles, layers, configuration, or artifacts.
  • Validate external input, authority, persistence, security, money, citations, and irreversible actions. Trust established internal invariants.
  • Unless a demonstrated requirement needs them, do not add hashes, UUIDs, random identifiers, retries, backoff, fallback chains, duplicate guards, speculative locks, compatibility layers, wrappers, factories, or placeholder interfaces.
  • Keep errors explicit. Do not hide defects behind broad catches, empty results, default values, or success-shaped degradation.
  • Use tokens by reducing uncertainty and repeated context: carry one canonical task contract, inspect unchanged inputs once, act when evidence is sufficient, and run one decisive check per claim. Do not hard-code token caps, truncate required context, or add token-shortening functions.
  • Record a reusable lesson only when reproduced evidence shows it generalizes. Scope the lesson narrowly; do not create ceremonial Skills, Wiki entries, or validation-only turns.
  • Review independently and read-only. Verify the changed user-visible path; broaden checks only when the blast radius is broad.
  • Report BLOCKED or UNTESTED honestly. Stop when the requested outcome is complete.
  • Unless the operator explicitly authorizes direct publication, provide a locally built Mac release package for the operator to install and try. Wait for explicit acceptance before pushing main, publishing a GitHub Release, or deploying the gateway. Do not replace the operator's running App automatically.

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.