Distillation

Date: Sep 16, 2026

• 🔴 [requested-validation] (22:35) User requested read-only exhaustive validation of immutable provider candidate /home/byk/Code/opencode-lore-provider-error, stable change qzwttkwltwzr, commit 493b0655a388, against base main@origin at 2cbf40060c67.
• 🔴 (22:35) User stated the expected plain cumulative diff SHA is 93617337752dee74776c2fb173d829f37c6616727fa72268f07c8c8f2f195422.
• 🔴 (22:35) User stated the candidate must contain exactly six specified files: 1. pipeline.ts, 2. stream/openai-responses.ts, 3. recall-aware test, 4. stream test, 5. pipeline-streaming.test.ts, 6. recall-codex-stream.test.ts.
• 🔴 (22:35) User stated the expected bundle hash is 9376719ec9b3af41ae9ae6154447be8a72c776540ec76773cacd861095399908.
• 🔴 (22:35) User stated a clean, empty child qnmstlol is expected.
• 🔴 [enforced-read-only-workflow] (22:35) User prohibited edits, formatting, dependency installation, VCS mutation, network access, server use, and production-database access during validation.
• 🔴 [requested-integrity-checks] (22:35) User required identity, hash, status, conflicts, and changed-file verification at both the start and end of validation.
• 🔴 [requested-tests] (22:35) User required the exact full-CI command pnpm test -- --printConsoleTrace with sufficient timeout; if the result is inconclusive, use 4 shards.
• 🔴 [requested-test-isolation] (22:35) User required verification of temporary-database isolation, explicitly excluding the production database.
• 🔴 [requested-artifact-validation] (22:35) User required verification of artifact hash/list, workers, final cleanliness, and stable hash.
• 🔴 (22:35) User required exact test counts, durations, exit codes, and an explicit PASS or FAIL.
• 🔴 (22:35) User directed that validation must never be warning-only; lore recall/UI fallback is required.
• 🔴 (22:36) User stated that in Jujutsu (jj), operations never fail on merge conflict; conflicts are recorded in the resulting commit and must be detected with jj st.
• 🔴 (22:36) User’s jj workflow rules: use jj rather than git when .jj/ exists; never use interactive flags such as -i/--interactive or jj resolve; always pass -m "msg" for describe/commit operations; verify mutations with jj st and jj log; prefer stable letter-based change IDs over hexadecimal commit IDs; never rebase or describe immutable commits such as main; use jj undo, jj op log, or jj op restore <op-id> for recovery.
• 🔴 (22:36) User stated tests use a temporary SQLite database through packages/core/test/setup.ts and never the production database.
• 🔴 (22:36) User stated releases use Sentry Craft via GitHub Actions and directed: never manually bump versions or edit CHANGELOG.md; Craft handles both automatically.
• 🔴 (22:36) User stated release triggering commands are gh workflow run release.yml -f version=auto for conventional-commit-derived semver or gh workflow run release.yml -f version=0.23.0 for an explicit version.
• 🔴 (22:39) User’s regression-test discipline requires a bug-fix regression test to fail on the base branch, proving reproduction, and pass on the fix.
• 🔴 (22:39) User stated tests that “test the guard” must fail when the guard is deleted.
• 🔴 (22:39) User directed: “Never trust the fix without seeing it fail.”
• 🔴 (22:39) User’s review workflow requires adversarial correctness review for every non-trivial PR; review evidence includes enumerated cases, apply-then-revert mutants, property-test stability across 10 runs, and an overall MERGE or DO-NOT-MERGE verdict rather than an unstructured “LGTM.”