API Tester

Generates API tests from real OpenAPI + route code, never guesses status codes

Tested · Works

Test report

Verdict
Tested · Works
Score
9.2/10
Tested
Jul 16, 2026
Environment
Claude Code 2.x (agent harness)
Upstream re-checked
Aug 10, 2026 · 756ad13

Assembled a real Express project with an intentional discrepancy between spec and code: openapi.yaml promised POST /users → 201, 400, and 401, but app.js actually returned 200 and 422, and 401 was unreachable (no auth-middleware or security scheme). Baseline set, generated from spec in the usual way, yielded 2 pass / 3 fail on `node --test`: expected 201 — got 200, expected 400 — got 422, expected 401 — got 200. SKILL.md set yielded 4 pass / 0 fail: the skill reads the implementation as the source of truth, asserts only truly reachable paths, and reports the three discrepancies instead of freezing them as falsely failing tests; it did not generate a test for 401 at all, following the rule 'do not assume every endpoint must return 400/401/403/404'. Stage 0 clean: a single SKILL.md without scripts, network, or exfiltration. The skill is entirely in Chinese — reports and test titles are in Chinese, which is friction for an English-speaking team; there is no automation in it, it's a disciplinary checklist.

Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works

  • Installs cleanly 5/5
  • Triggers reliably 5/5
  • Output vs. baseline 9/10
  • Docs & honesty 4/5

What API Tester does

Generates and runs API tests using only verifiable contract evidence — the OpenAPI spec, route/handler implementation, DTOs and existing tests — and records spec-vs-code divergences instead of freezing guesses into assertions. Triggers when you ask for integration tests, contract tests, regression cases for an endpoint, or API compatibility checks. Separates 'generate tests' from 'execute requests' and refuses write requests against unconfirmed environments.

How to install API Tester

git clone https://github.com/laolaoshiren/claude-code-skills-zh.git
mkdir -p ~/.claude/skills
cp -r claude-code-skills-zh/skills/api-tester ~/.claude/skills/api-tester

Skills live in ~/.claude/skills/ (global) or .claude/skills/ (per-project). Restart Claude Code after installing.

Commands — how to trigger API Tester

  • /api-tester Generates API tests from real OpenAPI + route code, never guesses status codes

It also activates on plain-language prompts like these:

  • Generate integration tests for this endpoint from the actual route code
  • Check if my OpenAPI spec matches what this handler actually returns
  • Write contract tests but flag any spec versus implementation mismatches

Frequently asked questions

Is the API Tester skill free?
Yes. The skill itself is free from laolaoshiren/claude-code-skills-zh. SkillProof publishes the install command and an independent test verdict at no cost.
Does API Tester work with Claude Code?
We tested it with Claude Code 2.x (agent harness) on Jul 16, 2026. Verdict: Tested · Works. Assembled a real Express project with an intentional discrepancy between spec and code: openapi.yaml promised POST /users → 201, 400, and 401, but app.js actually returned 200 and 422, and 401 was unreachable (no auth-middleware or security scheme). Baseline set, generated from spec in the usual way, yielded 2 pass / 3 fail on `node --test`: expected 201 — got 200, expected 400 — got 422, expected 401 — got 200. SKILL.md set yielded 4 pass / 0 fail: the skill reads the implementation as the source of truth, asserts only truly reachable paths, and reports the three discrepancies instead of freezing them as falsely failing tests; it did not generate a test for 401 at all, following the rule 'do not assume every endpoint must return 400/401/403/404'. Stage 0 clean: a single SKILL.md without scripts, network, or exfiltration. The skill is entirely in Chinese — reports and test titles are in Chinese, which is friction for an English-speaking team; there is no automation in it, it's a disciplinary checklist.
What is the API Tester SkillProof Score?
9.2/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 9/10, docs & honesty 4/5.
How do I install API Tester?
Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
Can I use API Tester with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.