First-5-minute kata
~15 minutes to complete your first scoped mission loop. Every step has a headless variant for CI and IDE agents.
Prerequisites: Node.js 24+, empty Git repository (or use a throwaway branch).
Steps
| # | Goal | Interactive | Headless |
|---|---|---|---|
| 1 | Install CLI | npm install -g @jeger-ai/opengantry | npx @jeger-ai/opengantry --version |
| 2 | Scaffold substrate | gantry init --tutorial | gantry init --yes --no-ci |
| 3 | Planner identity | gantry planner set "$(git config user.email)" | same (requires git config) |
| 4 | Health check | gantry onboarding | gantry doctor --json → exit 0 |
| 5 | Legislate mission | gantry start "First kata change" --msn MSN-KATA --skill-key logic --gate-command "npm test" | gantry legislate "First kata change" --msn MSN-KATA --skill-key logic --gate-command "npm test" |
| 6 | Planner commit | Review YAML, then git commit -m "[MSN-KATA] legislate mission" | same |
| 7 | Executor env | eval "$(gantry runtime env --mission .gitagent/missions/MSN-KATA.*.yaml)" | gantry runtime env --mission … --json |
| 8 | Execute change | Edit within mission tmvc_roots | gantry runtime exec --mission … -- npm test (after edits) |
| 9 | Trace evidence | Append unique line to EXECUTOR_LOG.md | same |
| 10 | Verify | gantry verify --mission … | gantry verify --mission … --json |
| 11 | Audit grep | git log --grep='MSN-KATA' --oneline | same |
Replace MSN-KATA with a valid id (e.g. MSN-0001) and match the legislated mission filename.
Expected outcomes
gantry verifyexits 0 with gate PASS and trace rows PASS (or PENDING cleared after quotes committed).git log --grep='MSN-'shows at least one legislative commit from a Planner allowlisted email.gantry status --jsonreports pinned or resolved mission context.
Friction log (optional)
When something fails in the first five minutes, capture:
step: <number>
expected: <what should happen>
actual: <what happened>
error_code: <GXT_* from --json if any>
File an issue or append to your team runbook — this is the highest-signal adoption feedback.
Related
- Operational runbook:
ADOPTION.md - Contrast specimens:
examples/contrast-agent-script/·examples/gantry-minimal/ - Time-to-Scaffold benchmark:
examples/benchmark-agent/(npm run examples:benchmark) · maintainer JSON:scripts/benchmark-scaffold.sh - IDE wiring:
INTEGRATIONS.md