Red team review
Skill richfrem/agent-plugins-skills/plugins/agent-loops/skills/red-team-review
repo for reusable plugins and skills
npx -y skills add richfrem/agent-plugins-skills --skill red-team-reviewAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 4 stars4 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
(Industry standard: Review and Critique Pattern) Primary Use Case: Iterative generation paired with adversarial review, continuing until an 'Approved' verdict is reached. Orchestrated adversarial review loop. Use when: research, designs, architectures, or decisions need to be reviewed by red team agents (human, browser, or CLI). Iterates in rounds of research → bundle → review → feedback until approved.
SKILL.md
4.1 KB, as published. Nobody here has run it
Red Team Review Loop
An iterative review loop where research is bundled via context-bundler and dispatched to one or more adversarial reviewers. The loop continues until the red team approves.
When to Use
- Architecture or design decisions that need adversarial scrutiny
- Research findings that need epistemic validation
- Security analysis that needs independent verification
- Any work product where "more eyes" reduce risk
Process Flow
- Research & Analyze — Deep-dive into the problem domain. Create analysis docs, capture sources.
- Review Packet Generation — Prepare the context for the reviewer:
- Create Prompt: Write or update a
red-team-prompt.mdexplaining exactly what is being reviewed and what the reviewer should focus on. - Define Manifest: Update a
manifest.jsonor equivalent list dictating which source files and research artifacts to include. - Bundle Context: Execute the
context-bundlerplugin, feeding it the manifest and prompt, to compile a single cohesive review packet. - Iteration Directory Isolation: Bundle the context and save the output to explicitly isolated directories (e.g.,
.history/review-iteration-1/) so that when the Red Team forces a rewrite, the baseline artifact is never destructively overwritten. 2.5. Interactively Determine CLI and Model (ask once during bootstrap): Before dispatching context bundles to CLI agents: - Interactively ask the user: "Which LLM CLI backend should be used for the adversarial review?" (Options:
agy,claude,copilot,codex,llama). - Ask: "Which specific model should be used?" (Present defaults, e.g.,
Claude Opus 4.6 (Thinking)for high reasoning orGemini 3.5 Flash (Low)for fast scans). - Ensure you append
< /dev/nullto the run command to preventSIGTTINhangs in headless execution engines.
- Create Prompt: Write or update a
- Dispatch to Reviewers — Send the bundle using the selected CLI and model to:
- Human reviewers (paste-to-chat or browser)
- CLI agents with adversarial personas (security auditor, devil's advocate)
- Browser-based agents for interactive review
- Receive Feedback — Capture the red team's verdict:
- "More Research Needed" → Loop back to step 1 with targeted questions
- Asynchronous Benchmark Metric Capture: Explicitly log the
total_tokensandduration_msused by the adversarial agent during this specific iteration into anevals/timing.jsonfile to calculate the true cost of approval. 4.5. Trust But Verify & TDD (Verification Gate): Do not blindly trust the reviewer's approval or feedback: - TDD Enforcement: Prioritize running unit and integration tests to ensure no regressions were introduced by any accepted recommendations.
- Delta Inspection: Check modified files directly for stubs, stales, or placeholders.
- Verify Critic Quality: Verify that the critic model's feedback is comprehensive and is not simply agreeing without actual critique.
- Completion & Handoff — Once the Red Team verdicts "Approved":
- Terminate the review loop.
- Pass the final, approved research and feedback documents back to the Orchestrator.
- DO NOT attempt to seal the session or run a retrospective. The Orchestrator handles that.
Dependencies
context-bundler— Required for creating review packets- Adversarial personas: user-supplied system prompt, or from an installed CLI agent plugin
(e.g., agent-personas). The
personas/directory is no longer bundled with agent-loops.