mission.md

Mission - Initialization Scripts for Agents

Goal

Build init_agent.py that probes runtime, dependencies, test command, env vars, and state freshness, then writes init_report.json and halts the session loud when a block-severity probe fails.

Inputs

• A repo with a requirements.txt (or equivalent), a test command, and the workbench state file from lesson 34
• The probe table from the lesson (runtime, deps, paths, env, state freshness, last-known-good commit)

Deliverables

• init_agent.py with one function per probe returning (name, status, detail)
• init_report.json carrying the full probe set and a timestamp
• Non-zero exit on any block-severity probe failure

Acceptance

• python3 code/main.py exits zero on the happy path
• Running it twice in a row is a no-op except for the timestamp
• A simulated missing env var probe surfaces in the report and flips the exit code

Out of scope

• Auto-installing missing dependencies. The script halts and surfaces; the human fixes.
• Calling an LLM from a probe. Probes stay deterministic plumbing.

References

• docs/en.md - full lesson
• code/main.py - reference implementation
• outputs/skill-init-script.md - extracted skill