data2dsl

Ticket 044: Koru Remediation Intent Generator (src/data2dsl_remediation.py and CLI feed-koru)

• ID: ticket-044
• Owner: unresolved:human
• Status: DONE
• Workflow state: DONE
• Created: 2026-08-25
• Closed: 2026-08-25
• Receipt: Implemented and validated locally.

Goal and scope

Implement RemediationIntentFormatter and format_remediation_intent() in src/data2dsl_remediation.py to produce structured remediation-intent/v1 payloads for semcod/koru closed-loop self-healing based on docs/research-koru-closed-loop.md:

1. Implement src/data2dsl_remediation.py transforming comparison results into actionable repair manifests.
2. Support deterministic status mapping (PROPOSED, SATISFIED, BLOCKED), typed action items (synchronize_metric, restore_missing_entries, resolve_conflict), and cryptographically pinned evidence digests.
3. Expose CLI command feed-koru in src/data2dsl_cli.py.
4. Provide unit test suite in tests/test_remediation_feed.py.

Acceptance criteria

• AC-01: src/data2dsl_remediation.py transforms comparison bundles into structured new-project.remediation-intent/v1 documents.
• AC-02: Actionable items and evidence digests are deterministically formatted for CONFLICT, MISSING_LEFT, MISSING_RIGHT, MATCH, and UNEVALUABLE.
• AC-03: CLI command feed-koru is implemented in src/data2dsl_cli.py.
• AC-04: Full unit test coverage in tests/test_remediation_feed.py.
• AC-05: The deterministic governance gate passes (GOV-PASS).

Participants

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