data2dsl

Ticket 079: Prevent identity fields from matching bottleneck queries

• ID: ticket-079
• Owner: founder:tom-sapletta-com
• Status: DONE
• Workflow state: DONE
• Created: 2026-08-30

Goal and scope

Repair the operational OR-query projection after live Supervisor integration showed that terms such as error also matched repository identity fields. List queries must select entity operational state, while the legacy single string query retains its full graph metadata search behavior.

Acceptance criteria

• AC-01: SESSION_EXECUTION_AUTHORIZATION is recorded from the Founder’s request to repair standardization regressions directly at their source.
• AC-02: A list query cannot select an entity solely through repository, ticket, pull-request, label or pointer identity.
• AC-03: The same list query selects entities through bounded operational status/error attributes and retains their immediate graph context.
• AC-04: A legacy string query still searches full node metadata.
• AC-05: Full tests, Ruff and governance pass; live fleet + PR projection contains no identity-only false positives.

Validation evidence

• Regression fixture proves list terms error and failed do not select healthy repositories named error_page or serialize-error, while a legacy string query still finds serialize-error by identity.
• Live fleet + PR-controller canary from the integrated, installed wheel returned a digest-verified 11-node, 10-edge graph and zero identity-only false positives against the current 21-PR controller cycle.
• Complete suite: 145 passed; focused Ruff, governance and whitespace gates pass. The existing jsonschema deprecation warning remains unchanged.
• Protected Validator run 33305732678 approved exact head 907a19a8449456044a6f88e6c175e7d5808771bc; PR #73 merged it as 2c639c456d17bfbf7da41bd03f5ce62ffb8072bf.

Participants

• Human participant: unresolved; no user-* file was created by this script.
• Agent participant: ai-codex.md