Handbook

L3.2 — Production-repo slice

Status: Defined

Updated

Sub-level at a glance
Parent level L3 — Use-case slice
Autonomous unit Slice in a production repo on a branch
Distinct assay evidence L3.1 evidence + target repo not in sandbox allowlist
Grade today a

Canonical policy, grades a-d, and promotion criteria: Autonomy levels (Blueprints). Per-project scoring: Autonomy maturity framework (Blueprints).

Gates and evidence

All gates and core evidence of the parent level (L3 — Use-case slice) apply unchanged. Declaring this sub-level adds the distinct check:

  • All L3.1 evidence, with the target repo not in the sandbox allowlist (a real repo on a branch)

A run that fails the sub-level check may still be reported at the plain parent-level grain if the parent's evidence holds — sub-level claims are strictly additive.

Prompt pack

Execution rubric (repo path, nav-hidden): prompts/autonomy/sublevel-l3-2.md — workcell-skeleton prompt declaring this sub-level, its evidence-to-return shape, and human gates.

Worked case

Hypothetical case — worked-through narrative in Blueprints Example 9; no matching campaign run exists.

Field Value
Repo forge-lenses on branch feat/scan-banner (hypothetical)
Goal Same three-layer slice as L3.1 but against a production repo outside the sandbox allowlist
Evidence produced Pattern: runs/campaigns/<campaign>/<item>/run-*/machine/run.json with target_repo: forge-lenses, sublevel: L3.2
Assay pass rationale All L3.1 evidence plus L3.2 _check_sublevel: target_repo declared and _is_sandbox(target_repo) false
Assay fail example Same slice against sandbox/app → L3.2 evidence: target repo is in the sandbox allowlist: sandbox/app