Phase 2, Stage → Launch

Turn the build sheet into runnable artifacts, then let the founder launch with their own key. No script here touches the network or the key, the user runs launch.sh.

Workflow

1. Generate payloads.

   python3 scripts/payload_generator.py \
     --sheet ./my-agent/build-sheet.json --out-dir ./my-agent
   # -> ./my-agent/payloads/{01-environment,02-agent,03-session,04-kickoff}.json
   

   Agent toolset → always_allow; every MCP toolset → always_ask (baked into the agent payload's permission_policies).
2. Write the launch script.

   python3 scripts/launch_script_writer.py --out-dir ./my-agent
   

   launch.sh creates environment → agent → session → kickoff in order, chaining IDs, and resumes on re-run (each step skips if its *.id file exists). It reads $ANTHROPIC_API_KEY at runtime.
3. Validate before launch.

   python3 scripts/payload_validator.py --dir ./my-agent
   

   FAIL blocks, especially a key_leak finding. Fix and re-run.
4. Minimal key step (never in chat). Check the shell first:

   [ -n "$ANTHROPIC_API_KEY" ] && echo "key present" || echo "export ANTHROPIC_API_KEY=... first"
   

   Point the founder to platform.claude.com → API keys. Never print the key to chat, never write it to a file.
5. Launch + watch the first poll.

   export ANTHROPIC_API_KEY=...      # in their shell, not in chat
   ./my-agent/launch.sh
   

   Mark checkpoints with Console deep links. Then goal_state.py set --phase grade-iterate and advance.

Hard rules (API-key safety)