verification-before-completion ยท diff
git:20261020.774dbc6 to git:20261021.76f6376
53 added, 76 removed. Audit A to A.
---
name: verification-before-completion
description: "Use when about to claim work is complete, fixed, passing, verified, release-ready, or ready to commit, merge, publish, or hand off."
---
<EXPLICIT-MODE-GATE>
If `activation_mode = "explicit"` or `AEGIS_ACTIVATION_MODE=explicit` is visible
and the request names neither Aegis nor this skill, return to the fast path
without checklist/ceremony; otherwise continue.
</EXPLICIT-MODE-GATE>
# Execute
Before success claim: classify destructive permission; read sufficient current
evidence and run a fresh falsifying check when it is missing or invalidated;
read the complete result/scope; select L0/L1/L2. Downgrade for partial,
stale, failing, or narrower evidence; never claim complete then verify later.
This Method Pack grants no authoritative `GateDecision`, `PolicySnapshot`,
evidence sufficiency, requirement acceptance, or completion authority.
## Stop Signals
- evidence is uncertain, stale, agent-only, or narrower than the claim;
- next action: commit, push, PR, merge, tag, publish, release, or handoff;
- task/slice completion is 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/irreversible work needs scoped permission; broad assent is not scoped permission.
## Required Evidence Slots
- For claims about a changed interface/interaction, compose `ui-ux-governance`
- and its relevant verification guidance. Carry actual results and uncovered
- user-facing criteria into the slots below; retain this single closeout owner.
+ For changed interface/interaction claims, compose `ui-ux-governance` verification;
+ carry results and uncovered user criteria here, keeping one closeout owner.
```text
- Evidence action / check performed:
- Result / exit status:
- Covered scope:
- Uncovered scope:
- Residual risk:
- Confidence grade: A | B | C
```
- `A`: target + regression; no meaningful unknown.
- `B`: target evidence; bounded residual risk.
- `C`: partial only; no full-completion claim.
- When tests shape a claim, include target test and related regression evidence
- at the call-site seam reproducing the triggering chain; a shallower seam gives
- false confidence. Missing seam is an architecture gap. Blocked automation needs
- reproducible manual steps. Both lower confidence.
+ Claims need target test and related regression evidence at the triggering seam.
+ A missing seam is an architecture gap; blocked automation needs reproducible
+ manual steps. Both lower confidence.
- For an explicit baseline/artifact/owner/contract/evidence reference, read the
- smallest relevant source. Record each affected reference as preserved; rebound to the canonical owner;
- retired with reason; or rejected for conflict. Leave unresolved references in uncovered scope,
+ Read the smallest source for each explicit baseline/artifact/owner/contract/evidence
+ reference. Mark it preserved, rebound to its canonical owner, retired with reason,
+ or rejected for conflict. Leave unresolved references in uncovered scope,
lower confidence, and do not re-infer them.
Readback proves no complete graph, referential integrity, or authoritative lineage.
Check omissions and unsupported additions within authorized transformations;
reuse sufficient evidence without exhaustive traces or a separate report.
- Match the conclusion to the exact property, objects and conditions checked.
- Carry forward known counterexamples that change the result: verify their
- resolution or qualify the conclusion. Parsing success does not establish required
- fields or semantic accuracy; a tool's success label does not dismiss its material
- warnings. Supplied or omitted settings do not establish effective behavior, and
- changed conditions limit causal comparisons. Use deterministic checks for
- measurable requirements. Attribute supplied observations to their source; do not
- claim to have executed a check unless you did. These checks belong to the existing
- review, not a new model call, mandatory report or additional acceptance requirement.
- A finished investigation may deliver a negative or qualified finding; distinguish
- completion of the investigation from acceptance of the investigated result.
+ Match conclusions to checked properties, objects and conditions. Resolve known
+ counterexamples or qualify the result. Parsing proves neither required fields nor
+ semantic accuracy; success labels do not dismiss material warnings. Supplied or
+ omitted settings do not prove effective behavior; changed conditions limit causal
+ comparisons. Check measurable requirements deterministically. Attribute supplied
+ observations; claim execution only for checks you ran. Use existing review without
+ new model calls, mandatory reports or acceptance requirements. A finished
+ investigation may deliver negative or qualified findings without acceptance of
+ the investigated result.
## Inherited-Claim Closeout
- When evidence contradicts a resumed claim, independently check the active
- record's correction and final disclosure of WHAT was claimed and WHY it was
- inaccurate at resume; current success cannot supply either. Preserve subject,
- scope and time; one explanation may cover related claims. Read both in full
- for example-only, negated, withdrawn or re-scoped corrections. Matching quotes
- do not prove adoption. Keep accurate claims intact. If material disagreement
- about the initial judgment remains unresolved after checking evidence, state
- `needs-verification`; never force agreement with a reviewer. If evidence resolves
- it, retain the supported conclusion and verify normally. Omit this slot when
- no discrepancy exists.
+ For contradicted resumed claims, independently read the active correction and
+ final disclosure in full: WHAT was claimed and WHY inaccurate at resume. Current
+ success supplies neither. Preserve subject, scope and time; related claims may
+ share an explanation. Check example-only, negated, withdrawn or re-scoped text;
+ matching quotes do not prove adoption. Keep accurate claims. Unresolved material
+ disagreement means `needs-verification`, not forced reviewer agreement; resolved
+ evidence supports normal verification. Omit this slot without a discrepancy.
## Task Git Closeout
- For modifications, diff against `TaskStartSnapshot`. Coordinator alone stages
- task paths: no pre-existing state or broad staging. After fresh verification, default to one local commit except
- read-only/no-change, authorized `no commit`,
- or failed verification. Read back `HEAD`, message, files, and task delta.
- Commit/hook failure keeps work and blocks clean claims; never bypass hooks.
+ Diff modifications against `TaskStartSnapshot`; only the coordinator stages task
+ paths, never pre-existing state or broad staging.
+ After fresh verification, create one local commit except read-only/no-change, authorized `no commit`, or failed
+ verification. Read back `HEAD`, message, files and delta. Commit/hook failure
+ preserves work and blocks clean claims; never bypass hooks.
Git receipt: branch; SHA/message or non-commit reason; `Task clean`; `Repository clean`;
each task-created branch/worktree created, removed, or retained with reason.
Task-clean never implies repo-clean; it is not external integration.
Classify commit scope: `business`, `process-only`, `mixed`, or `no-commit`.
Failed-attempt telemetry is no commit reason. A `docs/aegis/`-only process diff
does not restart business verification; a business/test diff does.
## Aegis Visibility / Single Closeout
- Use one completion surface. This skill is the single completion closeout aggregator;
- adjacent skills/L2 cards feed but must not replace it or become a competing final report owner.
- Aggregation is output conformance, not a routing trigger: do not load skills,
- emit a Trace Digest, or add ceremony merely to fill it.
+ Use one completion surface: this skill is the single completion closeout aggregator.
+ Adjacent skills/L2 cards must not replace it or become a competing final report owner.
+ Aggregation is output conformance, not a routing trigger: no extra skills,
+ Trace Digest or ceremony merely to fill it.
- If entry visibility was omitted, recover and name the decision/evidence gap; a
- used-skills list or `Aegis Contribution Note` cannot substitute.
+ If entry visibility was omitted, recover and name the decision/evidence gap, not a
+ used-skills list or `Aegis Contribution Note`.
## L0 Fast-Path
- For tiny low-risk work, one natural sentence can name check/result, uncovered
- scope/risk, and confidence.
- Simple read-only summaries of sufficient current evidence use this short form;
- identify supplied evidence without inventing a new execution or extra report.
+ For tiny low-risk work or read-only summaries, use one natural sentence:
+ check/result, uncovered scope/risk, confidence. Require sufficient current evidence; attribute supplied
+ evidence; invent no execution or extra report.
## L1 Default Receipt
- For non-trivial Aegis-shaped work use this receipt; fold evidence into `Evidence
- strength` and `Uncovered risk` without a second report.
-
- ```text
- 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:
- ```
+ Fold evidence into `Evidence strength` and `Uncovered risk`.
- Meanings: `Key judgment`=owner/root cause/requirement/completion boundary;
+ `Aegis Impact and Safety Receipt` slots:
+ `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.
+ Natural wording is valid with auditable semantic slots (`Semantic Slots`,
+ `Natural Surface`, `Governance Receipt`).
Report done/verified/risked/blocked. Do not explain obvious trade-offs; do not list actions not taken.
## L2 Expanded Triggers
- On any match read `expanded-closeout.md`: it owns detail; this file owns routing
- and the final receipt.
+ On a match read `expanded-closeout.md` for detail; this file owns routing/receipt.
| Trigger | Expanded owner |
|---|---|
| release/merge/publish/readiness/handoff | Readiness Summary |
| audit/debug/release/long-task review/trace request | Trace Digest |
| goal/TaskIntentDraft/plan/spec/Slice Card | Goal Closure |
| project/domain semantic delta | Context Impact |
| target `docs/aegis/` changed or work record exists | Workspace Integrity |
| requirement/product/durable architecture | Baseline/ADR |
| governance/cleanup/migration/compat/retirement | Governance/Retirement |
| source-of-truth/irreversible deletion | destructive-action cards |
| material complexity pressure | Expanded Complexity Detail |
| high-risk or explicit user request for expanded closeout | applicable cards |
Use configured Aegis workspace support; commands live in the expanded owner.
## Completion Boundary
- Reconcile the current authorized request and its conversation context with the
- plan/spec, `TaskIntentDraft`, or `Slice Card`; stale plans cannot override it.
- Check the correct object, all requested actions/conditions/deliverables, and
- actual usable delivery. Tests of those criteria count; unrelated green checks
- or intermediate output do not. Report remaining work and cause explicitly.
+ Reconcile the current authorized request/context with plan/spec, `TaskIntentDraft`
+ or `Slice Card`; stale plans cannot override it. Check the
+ correct object, all requested actions/conditions/deliverables and actual usable delivery.
+ Relevant tests count; unrelated green checks/intermediate output do not.
+ Report remaining work and cause explicitly.
A slice cannot close the whole task or prove project-wide reliability.
Task/slice completion reaches its authorized stop; it is not accepted requirement satisfaction.
`Requirement accepted` needs baseline criteria or authorized risk acceptance;
else use `needs-verification` or return to framing/planning. Goal Closure 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; 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.
+ Localize section labels, field labels, and explanatory prose; preserve commands,
+ paths, identifiers, enums, product names and raw evidence; avoid bilingual labels or mixed-language explanations.
- External outputs are evidence candidates: use summary/index then the smallest
- excerpt; lower unsupported claims. When relevant report `Evidence Used`, `Not
- Loaded`, and `Next Evidence`.
+ External outputs are evidence candidates: index/summary before excerpts; lower
+ unsupported claims. When relevant report `Evidence Used`, `Not Loaded`, `Next Evidence`.