pydantic-ai-skills tested by Argusic Runner

What the test found

pydantic-ai-skills works against the current pydantic-ai-harness line: the package installs, all 351 tests pass, and a real agent run loads a bundled skill, reads a reference file, and executes its script, with Git and S3 registries syncing libraries harness accepts.

Real gateway model (gateway/openai:gpt-5.2) needs PYDANTIC_AI_GATEWAY_API_KEY which is absent; the documented load_capability -> read_skill_resource -> run_skill_script flow was driven end-to-end with FunctionModel against real on-disk skills. Git registry verified against a real local git repo; S3 registry against an in-memory boto3 mock; require_loaded gate verified. Wheel built and re-verified in a fresh venv on latest harness 0.60/slim 2.60. 12 pre-existing ruff findings in untouched files remain (SIM117/TRY004/RUF012/RUF009/S102/RUF022).

Argusic installed pydantic-ai-skills in 3 minutes and launched it; verification reached Verified running for real. 5 of 5 recorded errors were worked through (see the timeline below).

Runsall runs of pydantic-ai-skills

This summary is drawn from the agent's recorded report for this run. Every figure traces to the log and recording above; nothing here is authored.