Correct CR-DD-016 capability-observation semantics - #156
Merged
coreytshaffer merged 1 commit intoAug 11, 2026
Merged
Conversation
Post-merge (#155) read-only reconciliation against CR-DD-013's actual resolve_capability() code and this repo's own tracked triagecore.toml found the CR's causal account of its own motivating evidence was wrong: a declaration with a resolved model binding authorizes route consideration without any probe, so "no probe means local_only fails closed" does not hold in general. - Add a Corrective Amendment section naming what changed and why; CR-DD-013 remains the governing runtime semantics. - Problem Statement / Determination: replace "undiscoverable prerequisite for execution" with the actual gap -- the assurance role of probing (upgrade vs. override) is undocumented. - Motivating Evidence: keep Trial 001 (genuinely pre-declaration, pre-CR-DD-014) as the real no-declaration/no-probe case; withdraw the Aug-8 trial as evidence for probe-absence-caused blocking -- its capability resolved fully available, and the block traces to a sensitivity-driven human_handoff selection, not capability. - Invariant 4: replace the falsified "fail closed when not observed" statement with the actual asymmetry (declaration+binding is sufficient absent an observation; a fresh negative observation overrides it; only no-usable-evidence-at-all fails closed). - In Scope / Acceptance Criterion 1: require the future quickstart text to document the three-case contract, not a "Configured-not- Observed block" that doesn't describe how the code behaves. Docs-only correction to a requirements contract; grants no implementation authority. Status remains Proposed.
✅ Deploy Preview for poetic-quokka-0fd859 ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
This was referenced Aug 11, 2026
6 tasks
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.
What this is
A corrective amendment to the already-merged CR-DD-016 (#155). Not a new CR — no new
policy or design decision is proposed here. CR-DD-013 remains the governing runtime
semantics; this PR brings CR-DD-016's description of those semantics back into
agreement with what
resolve_capability()actually does.Why
Post-merge, read-only reconciliation against
resolve_capability()(
triage_core/capability_evidence.py) and this repository's own trackedtriagecore.tomlfound that CR-DD-016's causal account of its own motivating evidencewas factually wrong: a
[capability]declaration with a resolved model bindingauthorizes route consideration without any probe. The repo's own checked-in
triagecore.tomlis in exactly that state right now (declare_local_fast = true,declare_local_heavy = true, both with model bindings, no probe record path set), andtest_declaration_without_observation_is_configured_not_observedpinslocal_fast_available is Truefor exactly this case.What this PR does
Determination that were discovered wrong after CR-DD-016: Document capability-probe operator workflow discoverability #155 merged.
triage_core/istouched; no schema, CLI, routing, or capability-resolution behavior changes.
task_id 12fe348e-94fd-43da-8bb8-2aa61a3eb949) asevidence for a probe-absence-causes-blocking claim. Its own
runner_selectedrecordshows both routes fully available (
capability_state: configured, bindings resolvedfor
local_fastandlocal_heavy); the actual block traces to a separate,sensitivity-driven
human_handoffselection inchoose_resilience_route, unrelated tocapability evidence. Day-1 Trial 001 (which ran before CR-DD-014 added those
declarations) remains the genuine no-declaration/no-probe example.
future operator-documentation contract (In Scope item 1, Acceptance Criterion 1) to
describe the real three-case behavior instead of a "
Configured-not-Observedblock"that the code doesn't actually produce on its own.
Status: ProposedandImplementation authority: Not authorizedare unchanged. The ProvisionalImplementation Allowlist, Explicitly Out of Scope, Non-Goals, and Sequencing sections
are untouched.
Scope
One file, docs-only:
docs/change/requests/CR-DD-016-capability-probe-operator-workflow-discoverability.mdtriage_core/,docs/daily_driver_quickstart.md, anyschema, test, backlog, or changelog file.
This correction is deliberately its own PR, landing before any implementation
authorization is considered for CR-DD-016, rather than being folded into that future
implementation work.
Verification
git diff --statgit diff --checkgit status --shortMentry, no residueTests not run: no executable, schema, or test surface changed.
🤖 Generated with Claude Code