The request "harness-review" does not name a specific product, and the candidates span very different areas, so the closest match is a tool that reviews code or agent harnesses. Nous Research describes Hermes Agent as an open source agent harness 10, and Trail of Bits publishes a harness-writing skill for fuzzing harnesses 9, which is the most literal fit for "harness review". Which kind of harness you mean (agent harness, fuzzing harness, or code review of pull requests) decides the best pi
A question back: Do you mean a fuzzing harness, an agent harness, or a code review of a pull request?
- 1
- 2
- 3
OpenAIopenai.comverified and signed
Provides a code-review action for running a final review on a pull request, and the Codex babysit-pr skill polls PR review comments and CI state 1.
official:MCP server · mcp-extensionsAPI · OpenAI APISDK · openailibrary · OpenAICLI · openai-whisper+177 community tools built for it
for agents: code-review · cad.reviewForm · code-review-testing
Sources
- 1babysit-pr — openai/codexskills.sh · read 2026-10-29
- 9Trail of Bits: Security Research, Audits, and Toolstrailofbits.com · read 2026-10-29
- 10Nous Researchnousresearch.com · read 2026-10-29
For agents: the 10 actions behind this answer
code-review · code-review verified and signed
Run a final code review on a pull request
security-review · security-review verified and signed
AI-powered codebase security scanner that reasons about code like a security researcher — tracing data flows, understanding component interactions, and catching vulnerabilities that pattern-matching tools miss. Use this skill when asked to scan code for security vulnerabilities, find bugs, check for
security-review · security-review verified and signed
Security code review for vulnerabilities. Use when asked to "security review", "find vulnerabilities", "check for security issues", "audit security", "OWASP review", or review code for injection, XSS, authentication, authorization, cryptography issues. Provides systematic review with confidence-base
mantis-review · mantis-review verified and signed
Independently reviews findings and filters out false positives. Use when consolidated findings need validation against the actual source code. Don't use for reproducing crashes or patching code.
PostReviewsReviewApprove · Stripe API verified and signed
Approve a review
dashboard-review · dashboard-review verified and signed
Use when reviewing dashboard JSON changes in a PR or branch. Extracts structured descriptions with kbdash, compares before/after, and checks guideline compliance.
create_pull_request_review · MCP-Brave-Search found in code, not yet checked
Create a review on a pull request
analyzing-release-readiness · analyzing-release-readiness verified and signed
Trigger a pre-merge release readiness review on a GitHub PR, GitLab MR, or local branch. Use when the user wants to analyze code changes for risk, correctness, and potential rollback issues before merging. Trigger words include release readiness, analyze PR, analyze MR, review PR, risk analysis, pre
harness-writing · harness-writing verified and signed
Designs and improves fuzzing harnesses for C/C++ and Rust. Covers mapping raw bytes onto a target API, generating structured inputs, avoiding non-determinism and false crashes, and deciding what to fuzz together. Use when writing a first LLVMFuzzerTestOneInput or fuzz_target! harness, when a campaig
hermes-agent · hermes-agent verified and signed
Use, configure, theme, extend, and orchestrate Hermes Agent.
Labels are the instruments' verdict and are never changed by ranking. Agents and robots get this same answer as JSON at /api/v1/find?q=…&answer=1 and through the MCP server.