agentleFS
theDRElabstheDRElabs/opencode-config/skills/tdd/SKILL.md

tdd

Skill1 starsChanged 3 months agoMITView on GitHub

Enforce observable red, green, and refactor evidence for one bounded issue before implementation is reported complete.

194 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use capture-learning-tools, design-frontend-dev 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.
  • The-Vibe-CompanySkill

    capture-learning-tools

    Turn technical conversations, bugs, CI failures, review comments,

    Also in 5 projects that use this

  • The-Vibe-CompanySkill

    design-frontend-dev

    Clean Impeccable frontend design skill. Use when the user wants to

    Also in 5 projects that use this

  • tristanphvnSkill

    ketchup-video

    Catch the user up like /keel:ketchup, with a short silent video of slides for what happened and the waiting calls as text under it. Use for /keel:ketchup-video.

    Also in 14 projects that use this

  • jaydubya818Skill

    figure-it-out

    Use when novel or multi-phase work needs an auditable workflow and no narrower playbook fits.

    Also in 54 projects that use this

  • jaydubya818Skill

    principle-laziness-protocol

    Apply when a refactor is adding wrappers or abstractions instead of simplifying.

    Also in 45 projects that use this

  • +2 more from The-Vibe-Company
  • +5 more from jaydubya818

The file

We split this skill into the rules it gives an agent

7 rules. 0 are common, 7 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 186 edited versions we read keep these.

  • 7

    Distinctive

    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:

    Use this skill only for one complete, human-accepted bounded issue.

    line 8
  2. Distinctive:

    Before inspecting project code, verify that the supplied issue artifact contains:

    line 14
  3. Distinctive:

    no unresolved human decision, unsafe HITL work, credential requirement, irreversible operation, or unknown command.

    line 20
  4. Distinctive:

    Red: add or strengthen the smallest acceptance-focused test, run the declared focused test command, and record a genuine non-zero result …

    line 39
  5. Distinctive:

    If behavior is genuinely not executable-testable, state why, name the substitute evidence, and obtain the issue's explicit human gate.

    line 55
  6. Distinctive:

    Run the project's declared fast checks sequentially, not invented commands.

    line 61
  7. Distinctive:

    every check's command, cwd, environment, duration, exit code, result, and complete evidence path.

    line 72

“Common” is measured: kept by at least 6 in 10 of the 186 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

TDD

Use this skill only for one complete, human-accepted bounded issue. The issue is the scope boundary; do not turn an implementation request into planning, backlog generation, architecture work, or unrelated cleanup.

Entry Gate

Before inspecting project code, verify that the supplied issue artifact contains:

  • one issue ID and one observable outcome;
  • explicit acceptance criteria, layers, modules, tests, constraints, and non-goals;
  • known project commands for every applicable fast check;
  • no unresolved human decision, unsafe HITL work, credential requirement, irreversible operation, or unknown command.

Reject or block the issue when any item is absent. TYPE: hitl, STATUS: blocked, COMMANDS: UNKNOWN, unresolved HITL, or unsafe work is never implementable by this agent. Do not infer missing details.

Relevant Surface

Discover project-local instructions and the declared commands first. Inspect only the modules, tests, migration files, and configuration needed by the issue. Treat unrelated existing changes as read-only and preserve them. Do not add public modules or cross boundaries not named by the issue; stop and report a boundary conflict instead of expanding scope.

Red, Green, Refactor

For behavior that can be tested:

  1. Red: add or strengthen the smallest acceptance-focused test, run the declared focused test command, and record a genuine non-zero result caused by the missing behavior. A syntax error, missing dependency, or broken setup is not red evidence.
  2. Green: make the smallest implementation that satisfies the acceptance criteria, rerun the same test, and record the zero exit result.
  3. Refactor: inspect the diff and simplify names, duplication, or structure without changing behavior; rerun the affected test and fast checks, recording the final result. Refactor evidence may be “no safe refactor needed” only when the diff review and final checks are recorded.

For migrations, red must be a failing migration/application assertion or rehearsal against a disposable database, green must show the migration and affected behavior passing, and refactor must rerun the migration checks after mechanical cleanup. Never run against production data.

If behavior is genuinely not executable-testable, state why, name the substitute evidence, and obtain the issue's explicit human gate. Do not claim TDD completion for unsafe or visual-taste-dependent work.

Checks and Evidence

Run the project's declared fast checks sequentially, not invented commands. Stop on the first blocking failure. Record each command exactly, cwd, command-scoped environment and inherited environment provenance, start/end or duration, exit code, and complete log or artifact path. FAIL blocks completion; BLOCKED and UNAVAILABLE require owner and residual-risk records.

Return an implementation-result artifact, not a completion assertion, containing:

  • issue ID and outcome;
  • files changed and tests added;
  • red, green, and refactor evidence;
  • every check's command, cwd, environment, duration, exit code, result, and complete evidence path;
  • preserved unrelated changes and module-boundary notes;
  • unresolved risks, unavailable checks, and any follow-up issue.

Implementation, project checks, independent review, and human acceptance are separate responsibilities. This skill never declares independent verification.

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.