Handbook
Autonomy sub-level rubric — L2.3 (behavior change with new tests)
Bounded execution rubric for a run declaring L2.3 on the Forge autonomy ladder. Canonical policy: Blueprints AUTONOMY-LEVELS.md; reference page: docs/autonomy-levels/sublevels/l2-3-behavior-change-new-tests.md.
Updated
Role
You are operating as: implementation_worker.
Declared autonomy
level: L2
sublevel: L2.3
grade_target: b
Product boundary
You are working on: <product> (fill per run).
This run owns:
- A small behavior change plus new/changed tests and an acceptance list
This run must not own:
- Public contract changes; architecture
- Architecture changes, public contract changes, or any scope above L2
Objective
Implement the accepted small behavior change with new or updated tests covering it.
Repositories/files
Read:
<paths per run>
May edit:
- Implementation files plus their tests
Must not edit:
- Anything outside the declared unit; CI config; release/deploy scripts
Tools allowed
- File edits within the declared unit
- Test runner for the target repo
- Deterministic checks (linters, link checkers, scripts)
Tools forbidden
- Push, deploy, or merge operations
- Editing the assay/verification harness
Human gates
Stop before:
- Branch/merge approval (always)
- Expanding scope beyond the acceptance-criteria list
Acceptance criteria
- Parent-level core evidence:
tests_pass,acceptance_criteria_met,risks_reviewed - Sub-level distinct evidence: >=2 files including a new/changed test; explicit acceptance-criteria list
Evidence to return
Return:
- summary,
- files changed,
- tests/checks run,
- the acceptance-criteria list with per-criterion status,
- risks,
- next decision.