Fix authenticated evaluator artifact JSON readback (#424)

[proerror77]

Change contract

Normalize an authenticated evaluator artifact through the same canonical JSON typed representation used by durable readback before sealing its digest, so a valid artifact remains valid after content-addressed persistence.

Closes #424.

Out of scope

Metric formula changes, explicit metric rounding, repository-wide canonical JSON changes, the #325 three-event fixture, collectors, runtime/deployment, CEX, alpha-harness, and hft-search-kernel.

Dependency / merge order

Base: main at 9d502944eede58508251b31714e58a8c78b10f55.

Merge this PR before rebasing/resuming #325. The original #325 three-task E2E is the dependent post-merge acceptance check.

Focused validation

• RED before fix, twice: cargo test --locked -p ploy-research --features db evaluator_artifact_digest_survives_canonical_json_readback -- --nocapture failed with held-out evaluator artifact digest mismatch.
• GREEN after fix: the same test passes through production content-addressed write, bounded verified readback, typed artifact validation, and result receipt validation.
• cargo test --locked -p ploy-research --features db --lib: 357 passed, 1 database-dependent test ignored.
• rustfmt --edition 2021 --check crates/ploy-research/src/prediction_mcts_authenticated.rs: passed.
• git diff --check: passed.
• cargo clippy --locked -p ploy-research --features db --lib --no-deps -- -D warnings -A clippy::type_complexity: passed. The exact allow covers two unchanged pre-existing lib findings; unscoped --all-targets remains blocked by unrelated existing warnings in examples and binaries.
• Matt code-review exact range 9d502944...4443c565: Standards 0 findings; Spec 0 findings.

A PRD is unnecessary because #424 is a small, already-specified defect with one trust-boundary behavior and one deterministic regression test.

Rollout / rollback impact

Research artifact identity only. New artifacts receive the synchronized policy identity sha256:76d4afe04854587a2883f06cdcf8a67563853f320630eafb700ef794b5ef9c33; existing mission inputs with the prior policy ID fail closed. Rollback is the single commit and its mechanical BTC/SOL policy digest sync. No live collector or execution state changes.

Summary by CodeRabbit

• Bug Fixes

  • Improved consistency when creating and validating authenticated prediction results by ensuring their recorded content remains canonical after being saved and reloaded.
  • Added validation coverage to confirm authenticated artifacts retain valid, matching integrity records throughout the readback process.

• Configuration

  • Updated the search policy snapshots used by the Bitcoin and Solana five-minute prediction-market research configurations.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: bcb4f3d0-ed27-4f74-a6b0-d75a50baa37e

📥 Commits

Reviewing files that changed from the base of the PR and between 9d50294 and 4443c56.

📒 Files selected for processing (3)

• rust_hft/prediction-markets/config/research_missions/polymarket-btc-5m.example.json
• rust_hft/prediction-markets/config/research_missions/polymarket-sol-5m.example.json
• rust_hft/prediction-markets/crates/ploy-research/src/prediction_mcts_authenticated.rs

---

📝 Walkthrough

Walkthrough

The change normalizes authenticated evaluator artifacts through canonical JSON readback, adds a persistence and receipt-validation regression test, and updates pinned search policy snapshot identifiers in the BTC and SOL Polymarket mission examples.

Changes

Authenticated artifact digest stability

Layer / File(s)	Summary
Canonical artifact normalization and validation rust_hft/prediction-markets/crates/ploy-research/src/prediction_mcts_authenticated.rs	AuthenticatedEvaluationArtifact::new canonicalizes its value through JSON deserialization, while a regression test verifies digest preservation after verified readback and successful receipt validation.

Mission snapshot updates

Layer / File(s)	Summary
Pinned search policy snapshots rust_hft/prediction-markets/config/research_missions/polymarket-{btc,sol}-5m.example.json	The BTC and SOL mission examples use new search_policy_snapshot_id values.

Estimated code review effort: 3 (Moderate) | ~20 minutes

Possibly related PRs

• proerror77/monday#412: Modifies authenticated evaluation and result digest consistency in the same module.
• proerror77/monday#408: Changes authenticated evaluation artifact creation and canonical JSON behavior.
• proerror77/monday#387: Introduces related authenticated artifact and receipt logic in the same module.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title is concise and accurately names the main change: fixing authenticated evaluator artifact JSON readback.
Description check	✅ Passed	All required sections are present and substantive; only the scope-exception section is implied rather than explicitly labeled.
Linked Issues check	✅ Passed	Issue #424's digest-stability fix, regression test, and preserved readback/receipt validation are implemented.
Out of Scope Changes check	✅ Passed	The PR stays focused on artifact digest stability; the policy hash sync is directly tied to the new artifact identity.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch codex/authenticated-artifact-readback-424

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}