Borrowing it
Nothing to install: this file belongs to redhat-community-ai-tools/harness-eval. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/redhat-community-ai-tools/harness-eval/main/.cursor/commands/harness-review.mdgit clone --depth 1 https://github.com/redhat-community-ai-tools/harness-evalWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/redhat-community-ai-tools/harness-eval/harness-review)<a href="https://agentmods.dev/commands/redhat-community-ai-tools/harness-eval/harness-review"><img src="https://agentmods.dev/badge/commands/redhat-community-ai-tools/harness-eval/harness-review/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/redhat-community-ai-tools/harness-eval/harness-review"><img src="https://agentmods.dev/badge/commands/redhat-community-ai-tools/harness-eval/harness-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.00643 |
| Opus 5 | $0.00000 | $0.00321 |
| Sonnet 5 | $0.00000 | $0.00129 |
| Haiku 4.5 | $0.00000 | $0.00064 |
Grade A, and why
harness-review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 64 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Eval Setup Review
Full qualitative review of the agent setup. Read every file, evaluate quality, redundancy, and optimization opportunities. Produce KEEP/REVIEW/REMOVE verdicts per component.
Hard Rules
- Never give a verdict without reading the files.
- Read before you judge. Read every file's actual content before assessing.
- Don't manufacture problems. If the setup is good, say so.
Step 1: Ask Output Preference
Ask the user: print the report in conversation, or write to a file?
Step 2: Run Lint for Context
Run the deterministic scan first to get structural data:
uvx --from harness-eval harness-eval harness-lint . --format json
If uvx is not available, fall back to pip install harness-eval and use harness-eval directly.
Read the JSON output. Use it as context for the qualitative review. Do NOT present the lint report separately.
Step 3: Read Actual Files
Read the actual content of every component discovered: .cursor/rules/*.mdc files, .cursorrules, .cursor/commands/*.md, skill SKILL.md files (including reference files in subdirectories), and .cursor/hooks.json.
Step 4: Evaluate Each Component
For each component, provide:
- Lint results: list each rule that failed and explain WHY it failed in one sentence
- A 2-3 sentence qualitative assessment (what it does, whether it adds value, whether it's well-built)
- Issues found, citing specific content
- Verdict: KEEP, REVIEW, or REMOVE
Evaluate across these areas:
- Specificity: does each component add value the AI doesn't already have? Would deleting it change behavior?
- Redundancy: does it duplicate content from other components or the AI's default behavior?
- Trigger quality: are descriptions specific enough to route correctly? Any overlapping triggers?
- Token efficiency: are components under 500 lines? Is content split between main file and references?
- Instruction clarity: contradictions, vague language, buried rules, orphaned conditionals?
- Impact: for each issue, what will go wrong at runtime if this isn't fixed?
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 64 lines · 0 tokens per session scan A 438eab503fe1
harness-review is a command published in the GitHub repository redhat-community-ai-tools/harness-eval (27 stars, last pushed 3d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 643 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
refactor-clean
You are a code refactoring expert specializing in clean code principles, SOLID design patterns, and modern software engineering best practices. Analyze and refactor the provided code to improve its quality, maintainability, and performance.
delegate-review
Run OCR in delegation mode — OCR handles file selection and rules, the host agent performs the actual review.
sdd-explore
Explore and investigate an idea or feature — reads codebase and compares approaches.
sdd-init
Initialize SDD context — detects project stack and bootstraps persistence backend.
gentle-sdd-ff
Fast-forward all SDD planning phases — proposal through tasks.
parallel-analyze
Multi-perspective analysis using parallel subagents. Analyzes code from architecture, security, performance, and testing viewpoints simultaneously.