feat(shadow-verify): merge forge-approved enhancements - #5
Merged
Conversation
…p, and gate verdict
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
This PR merges forge-approved enhancements into the
shadow-verifyskill, generated by/forgefrom pattern cardshadow-verify-should-be-invoked-as-a-structural-checkand approved after 3 qualify iterations (SALVAGE → SALVAGE → APPROVE, 29/30 Stage 1, 24/24 Stage 2).Changes
1. Claim normalization format
Before dispatching verifiers, the coordinating agent now normalizes each claim to a canonical form:
evidence_refmust be a concrete pointer (file path, line number, git ref, config key, or API endpoint). Claims that cannot be anchored to a concreteevidence_refare immediately classifiedUNVERIFIABLEand removed from the verification queue, with a dedicated output section explaining why each could not be anchored.2. Scale cap (50-item halt)
If the normalized claim list exceeds 50 items, the coordinating agent halts and asks the operator to scope the investigation before proceeding. Verification at scale degrades into noise; the cap prevents a finding flood from becoming an echo-chamber rubber-stamp.
3. STALE verdict tier
Added
STALEas a first-class verifier verdict alongsideCONFIRMED,REFUTED, andUNVERIFIABLE. ASTALEverdict means the claim was true at some prior point but no longer holds. The verifier emits a corrected/updated finding, and the claim is surfaced with[was true, now stale]— it must not be forwarded downstream without re-investigation if currency matters for the decision.4. PASS/WARN/FAIL gate verdict
After all verifier results are merged, the coordinating agent now emits a single aggregate gate verdict before surfacing findings downstream:
PASSWARNSTALE,UNVERIFIABLE, or composition/echo-chamber flagged; proceed with explicit caveatsFAILREFUTED; findings must not be forwarded until re-investigatedA
FAILverdict hard-blocks downstream use. AWARNmay proceed but must carry flagged claims and their tags — stripping caveats before forwarding is a protocol violation.5. Gap-analysis coverage in trigger scope
The
When to invokesection now explicitly includes gap analyses, audit findings, and issue lists that will be forwarded to a downstream session as trigger conditions, alongside the existing code reviews, audits, findings reports, and large refactors.Provenance
~/.afk/agent-framework/briefs/merge-pending/shadow-verify-2026-08-24T02-08-44Z.mdshadow-verify-should-be-invoked-as-a-structural-check