test-author
Skill1 starsChanged 46 days agoMITView on GitHub
Write a new unit test or E2E flow under this project's Testing Policy when you cannot dispatch the unit-test-author / e2e-test-author agent (no Agent tool). Invoke before adding any new test file or test case. Args - unit | e2e.
4 other projects use a version of this.
Adopted by
Most recently taken up by
- hams-ollo/zen-agent-skillstheir own version, first seen Oct 20, 2026
- jpicklyk/task-orchestratortheir own version, first seen Oct 17, 2026
- emtcmca/guildproof-skillstheir own version, first seen Sep 19, 2026
- rohitgollarahalli/claude-skillstheir own version, first seen Jun 22, 2026
Other versions4 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 4
- jpicklyk/task-orchestrator★ 214same name
- hams-ollo/zen-agent-skills★ 2same name
- emtcmca/guildproof-skills★ 0same name
- rohitgollarahalli/claude-skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository15 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- discover-setupSkill
- discoverSkill
- discussSkill
- do-code-reviewSkill
- doSkill
- improve-promptSkill
- journeySkill
- prototypeSkill
- setup-pre-commitSkill
- sketchSkill
- specSkill
- testing-policySkill
- test-triageSkill
- ticketsSkill
The file
We split this skill into the rules it gives an agent
1 rule. 0 are common, 1 is distinctive, and 0 are very specific to this project.
0
Common
Most of the 4 edited versions we read keep these.
1
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
- Distinctive: line 14
## Dispatch protocol: does not apply inline, since you are the caller: you own the feature code, the stack and the commit.
“Common” is measured: kept by at least 6 in 10 of the 4 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.
Files
<!-- testing-policy:skill v=3.0 -->You are about to write a new test inline, without the test-author agent. The agent file is the single source of the rules; this skill only routes you to the parts that bind inline.
- Fill the dispatch input for yourself first, for the one test you are about to write: behavior to prove (it becomes the test name), who relies on it and what a wrong or missing result costs them, target (unit) or journey/screen (E2E), origin (
bugfix/new feature), expected red (unit) or fixture state (E2E). If you cannot, you do not yet know what you are testing: stop and ask. A behavior no one relies on pins structure: it gets no test. One test per pass: the next input is written after this test is green and its implementation exists, never a batch ahead of the code. Open the agent file and apply ## Authoring rules and ## Project map in full:
unit→.claude/agents/unit-test-author.mde2e→ this repo has no E2E flows: the unit test is the whole proof
## Dispatch protocol does not apply inline, since you are the caller: you own the feature code, the stack and the commit.
- Run the Discovery block from Project map before writing anything, and keep what it returned.
- Write and run as the rules say. Unit: red for the declared reason, then only the code this test needs, then green, then refactor on green with the suite rerun; E2E: preflight, then green against the real stack. Never weaken an assertion to get there.
- Report inline the same three sections the agent would: Outcome (the assertion that proves the behavior,
file:line, and on E2E each settle point with its anchor), Reuse audit (searches run, what they returned, decision per asset; present even when nothing was created) and Run (command and output verbatim). A promotion is listed as its own changeset and lands in the same commit as the test, or in a refactor commit immediately before it.
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
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.

