Adonis0123/adonis-skills / skills/open-code-review-loop
open-code-review-loop
UnverifiedRun a bounded OCR-delegation review-fix-re-review loop on a Git workspace until current evidence validates NO_FINDINGS. Use only when the user wants verified findings fixed and independently re-reviewed until clean with codex, claude-code, grok-build, or cursor-cli. Do not use for one-shot open-code-review-delegate or read-only review. A full loop request asks at most once for an unnamed Reviewer (recommending a product other than the host), then runs host mutation checks without further confirmation. Fail closed on missing ocr, reviewer mutation, skipped files, malformed output, stale evidence, or exhausted rounds.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 19.3 KB |
| evals/evals.json | config | 20.8 KB |
| evals/trigger-eval.json | config | 2.5 KB |
| references/adapter-contracts.md | markdown | 13.5 KB |
| references/fix-contract.md | markdown | 2.7 KB |
| references/fix-schema.json | config | 857 B |
| references/review-contract.md | markdown | 3.1 KB |
| references/review-schema.json | config | 2.8 KB |
| scripts/build_review_bundle.pyexecutable | script | 27.0 KB |
| scripts/extract_product_output.py | script | 3.6 KB |
| scripts/test_extract_product_output.py | script | 6.2 KB |
| scripts/test_validate_fix_result.py | script | 5.1 KB |
| scripts/test_validate_round.py | script | 103.0 KB |
| scripts/validate_fix_result.py | script | 5.4 KB |
| scripts/validate_round.pyexecutable | script | 41.4 KB |
references/review-contract.md
View on GitHub# Reviewer result contract
Load this reference before invoking the selected Reviewer. The Reviewer must
return one JSON object and no surrounding Markdown. Use
[review-schema.json](review-schema.json) when the product supports an output
schema.
```json
{
"schema_version": "1",
"evidence_id": "sha256:<hex>",
"reviewer": {
"product": "codex|claude-code|grok-build|cursor-cli",
"session_id": null
},
"files": [
{
"path": "relative/path.ts",
"status": "modified",
"disposition": "reviewed|skipped",
"reason": "review rationale when reviewed; required when skipped"
}
],
"findings": [
{
"id": "OCR-001",
"path": "relative/path.ts",
"start_line": 10,
"end_line": 12,
"category": "bug|security|performance|maintainability|test|style|documentation|other",
"severity": "critical|high|medium|low",
"content": "Evidence-backed problem and impact",
"required_fix": "Smallest acceptable repair",
"acceptance_check": "Command or observable check that would falsify the problem"
}
],
"verdict": "FINDINGS|NO_FINDINGS|BLOCKED",
"block_reason": null
}
```
## Invariants
1. `evidence_id` must equal the supplied bundle ID.
2. The selected review set is the union of `reviewable_files` and
`supplemental_reviewable_files`. `files` must account for every selected
`(path, status)` identity exactly once, applying both `rules` and
`supplemental_rules`.
3. `NO_FINDINGS` requires an empty findings list, 100% reviewed coverage, and
zero skipped files. The host additionally rejects clean when the bundle has
any `unaccepted_excluded_files`.
4. `FINDINGS` requires at least one actionable finding and full file coverage.
5. `BLOCKED` requires a concrete reason. It is not a clean verdict.
6. Every finding path must belong to the selected review set. Nearby context may
support a finding but does not silently expand the writable scope.
7. Do not report preferences, speculative redesigns, or unverified external
claims. A low-severity item is still expected to be worth fixing now.
8. The Reviewer is read-only. It may run checks that require no writes, but must
not edit source, tests, configs, Git state, the bundle, or temporary paths.
The host runs tests that need caches or temporary directories.
9. The visible host replaces `reviewer.product` and `reviewer.session_id` with
the canonical selected product and the recorded product session. Never
trust model-authored identity fields.
The host validates this object with `scripts/validate_round.py`. One malformed
response may be corrected in the same Reviewer session; a second failure ends
the loop as `UNVERIFIED`.
`start_line` and `end_line` are positive integers when the issue maps to lines
in the new file. Use `null` for both on deleted files or a genuinely path-level
finding; never invent a line merely to satisfy the shape. Use `null` for
`block_reason` except on `BLOCKED`.
The portable provider schema keeps the two line fields independently nullable
because supported structured-output subsets differ. The local validator is the
authoritative semantic gate and rejects a one-null/one-integer pair.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |