OpenParable
Skip to content
← Back to skills

Chat2api Proxy Testing

Testing & Quality
chat2api-proxy-testing

Chat2api Proxy Testing is an Agent Skill for testing & quality published in the xiaoY233/Chat2API repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Use when validating Chat2API Manager proxy behavior across dialogue, tool calling, context, provider routing, request logs, and live client replay workflows.

View on GitHub ↗
QUICK START

How to use this skill

Bring this guide into your coding agent with a prompt tailored to the tool you use.

  1. Open your project in Codex.
  2. Copy and paste the prompt below.
  3. Review files and risks before approving any action.
Prompt to paste
Inspect the Agent Skill at https://github.com/xiaoY233/Chat2API/blob/HEAD/skills/chat2api-proxy-testing/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/chat2api-proxy-testing, but wait for my approval before writing files or running scripts.

Copying this prompt does not install or run the skill. Review third-party files before use.

Chat2API Proxy Testing

Use this as the orchestration entry point.

Focused Skills

  • Use chat2api-management-api for management API setup, observation, and cleanup.
  • Use chat2api-har-tool-fixture to extract reusable fixtures from HAR files.
  • Use chat2api-tool-client-replay to replay one client fixture against one or more models.
  • Use chat2api-provider-model-matrix to discover /v1/models and run matrix coverage.

Default Order

  1. Run source regressions when code changed.
  2. Snapshot health and config through management API.
  3. Extract or select a fixture.
  4. Discover models through /v1/models.
  5. Replay scenarios.
  6. Inspect logs and sessions.
  7. Restore config and delete temporary test assets.

Source Of Truth

The versioned source of truth is this skills/chat2api-proxy-testing directory plus the focused sibling skills. ignored .codex copies are local working copies only and must not be the only place where testing behavior is documented.