Verification Before Completion

Execute

Before any success claim, classify destructive permission needs, choose and run a fresh falsifying check, read its complete result/scope, then select L0/L1/L2. If evidence is partial, stale, failing, or narrower than the claim, downgrade; never claim complete first and verify later.

This Method Pack grants no authoritative GateDecision, PolicySnapshot, evidence sufficiency, requirement acceptance, or completion authority.

Stop Signals

Stop before claiming or advancing when:

• evidence is uncertain, stale, agent-only, or narrower than the claim;
• the next action is commit, push, PR, merge, tag, publish, release, or handoff;
• task/slice completion is being treated as accepted requirement satisfaction;
• governance or retirement lacks repair/retirement evidence;
• retained old logic lacks a retention reason and retirement trigger; or
• complexity closure is unresolved.

Destructive or irreversible work needs scoped permission; warnings or broad assent do not grant it.

Required Evidence Slots

Keep these slots explicit and auditable:

- Evidence action / check performed:
- Result / exit status:
- Covered scope:
- Uncovered scope:
- Residual risk:
- Confidence grade: A | B | C

• A: direct target plus relevant regression evidence; no meaningful unknown.
• B: direct target evidence with bounded residual risk.
• C: partial evidence only; do not claim full completion.

When tests shape the claim, include target test and related regression evidence. Place the regression test at the seam that exercises the real bug pattern as it occurs at the call site; a seam too shallow to replicate the triggering chain gives false confidence. If no such seam exists, record the missing seam as the finding (the architecture prevents locking the bug down) and lower confidence. If automation is blocked, give reproducible manual steps and lower confidence. Evidence is not completion authority.

When a completion claim depends on an explicit baseline, artifact, owner, contract, or evidence reference, read back the smallest relevant source. For affected known references, verify a disposition: preserved, rebound to the canonical owner, retired with reason, or rejected because of conflict. Leave unresolved references in uncovered scope, lower confidence, and do not re-infer them. This readback does not prove a complete relationship graph, referential integrity, or authoritative lineage.

Task Git Closeout

For modification tasks, compare the final state with TaskStartSnapshot. Only the coordinator stages task-owned paths; never use broad staging or include pre-existing user state. A default local task commit follows fresh verification unless the task is read-only/no-change, user/project authority says no commit, or verification failed. Read back HEAD, message, files, and remaining task delta. Commit/hook failure preserves the work and blocks a clean claim; do not bypass hooks.

The Git receipt reports branch; commit SHA/message or non-commit reason; Task clean; Repository clean; and each task-created branch/worktree as created, removed, or retained with reason. Task-clean never implies repo-clean. This receipt is evidence, not external integration or completion authority.

Classify commit scope before claiming verification stability: business, process-only, mixed, or no-commit. Failed attempt telemetry is not a commit reason. A diff limited to docs/aegis/ process records does not restart already completed business-code verification; a business or test diff does.

Aegis Visibility / Single Closeout

Use one completion surface; no parallel final reports. verification-before-completion is the single completion closeout aggregator. Adjacent skills and L2 cards feed the receipt but must not replace it or become a competing final report owner. Receipt aggregation is output conformance, not a routing trigger: do not load extra skills, emit a Trace Digest, or add ceremony merely to fill the receipt.

If entry visibility was omitted, recover the decision/evidence boundary and name the gap; a used-skills list or Aegis Contribution Note is no substitute.

L0 Fast-Path

For tiny low-risk work, one natural sentence can name check/result, uncovered scope/risk, and confidence.

L1 Default Receipt

For non-trivial Aegis-shaped work, use this receipt. Evidence slots fold into Evidence strength and Uncovered risk; avoid a second evidence report.

Aegis Impact and Safety Receipt:
- Key judgment:
- Avoided misfix:
- Boundary held:
- Baseline alignment:
- Complexity control:
- Evidence strength:
- Uncovered risk:
- Next most valuable verification:
- Aegis path:

Field meanings: Key judgment=owner/root cause/requirement/completion boundary; Avoided misfix=fallback/duplicate/test accommodation/scope growth; Boundary held=contract/owner/baseline/non-goal/data/runtime boundary; Baseline alignment=aligned/Design Defect/Implementation Drift/missing-authority/needs-clarification/not triggered; Complexity control=completion-time delta/closure; Evidence strength=fresh check/result/scope/confidence; Uncovered risk=remaining gaps/residual risk; Next most valuable verification=highest-value next check; Aegis path=optional, not judgment/evidence.

Natural wording is valid when every semantic slot stays auditable. Semantic Slots, Natural Surface, and Governance Receipt are compatibility names, not other reports.

Report what was done, verified, risked, and blocked. Do not explain obvious trade-offs and do not list actions not taken; silence on unexecuted options is the compact default, not an omission.

L2 Expanded Triggers

On any match, read expanded-closeout.md. It owns detail; this file owns routing and the final receipt.

For target workspace changes, keep configured Aegis workspace support wired. When a work record exists run python <aegis-workspace-helper> bundle --root <target-project-root> --work YYYY-MM-DD-<slug>, then run python <aegis-workspace-helper> check --root <target-project-root>. These checks prove structure, not evidence sufficiency.

Completion Boundary

Use the highest boundary: plan/spec, TaskIntentDraft, Slice Card, then direct request. Claim only what fresh evidence covers; slice evidence cannot close the whole task.

Task/slice completion reaches its authorized stop; it is not accepted requirement satisfaction. Requirement accepted needs baseline criteria or authorized risk acceptance. If unclear, use needs-verification or return to framing/planning.

Goal Closure stop states: done | blocked | needs-verification | scope-exceeded.

An Execution Readiness View is input, not verification evidence.

Complexity Downgrade

For non-trivial code, inspect the diff and use using-aegis/references/complexity-governance.md plus docs/current/AEGIS_COMPLEXITY_GOVERNANCE_BASELINE.md; emit one Complexity control line.

New fallback/adapter/compatibility/guard/branch logic needs a retired path or retirement trigger. Complexity Closure: exceeded-unresolved blocks completion. Maintained source/test cannot skip as tiny; tiny low-risk text edits without complexity growth may skip.

Output and Prompt Hygiene

Localize section labels, field labels, and explanatory prose. Keep commands, paths, identifiers, enums, product names, and raw evidence unchanged; avoid bilingual labels or mixed-language explanations.

External outputs are evidence candidates. Prefer summary/index and the smallest excerpt; lower unsupported claims. When relevant report Evidence Used, Not Loaded, and Next Evidence.