feat: derived policy_conformance rollup (guardrail-receipt surfacing, honest advisory seam) - #109
Merged
Merged
Conversation
…ail-receipt surfacing, honest advisory seam) Co-Authored-By: Codex (gpt-5.6-sol) <noreply@openai.com>
moonweave
added a commit
that referenced
this pull request
Jul 21, 2026
Ships the derived policy_conformance rollup (#109): a top-level guardrail verdict projecting the role-capability conformance axes, with the honest advisory seam. Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
First step of the 'guardrail receipt' product frame (2026-07-21 strategy review). Adds a derived, non-gating top-level rollup that answers 'did the agent stay in its declared lanes?' directly, projecting the existing role-capability conformance axes (write_scope/tool_calls/skill_routing).
Purely additive — no re-derivation, contract, schema, philosophy, or gating change.
policy_conformanceis computed from the already-computedrole_capability_conformance+ each axis's enforcement mode.The honest advisory seam (the one place
decision:passcan coexist with a real violation): an advisory skill_routing violation yieldspolicy_conformance.overall == failwith the axis markedenforcement: advisory, blocks_handoff: false— so the guardrail question is answered truthfully (it left its lane) AND it is explicit that it did not block handoff. It never silently contradictsdecision.overallisinconclusivewhen a required axis was undeclared ('claimed a guardrail but didn't install it'), and the field isNone(byte-stable) when there are no policy axes.Verification (independent, Opus): full suite 457, only the 1 pre-existing failure, 0 new. Live-proved the advisory seam (policy fail + blocks_handoff false) and the block case (fail + blocks_handoff true) and the None case. Mutation-checked (making
overallignore advisory violations reddens the advisory test; write_scope unaffected). Human render surfaces the honest line.🤖 Generated with Claude Code