Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add vsladkov/claudex-stereo/plugin install stereoWrote 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/agents/vsladkov/claudex-stereo/reviewer)<a href="https://agentmods.dev/agents/vsladkov/claudex-stereo/reviewer"><img src="https://agentmods.dev/badge/agents/vsladkov/claudex-stereo/reviewer.svg" alt="Measured on agentmods" 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 | $0.00018 | $0.00385 |
| Opus 5 | $0.00009 | $0.00192 |
| Sonnet 5 | $0.00004 | $0.00077 |
| Haiku 4.5 | $0.00002 | $0.00038 |
Grade A, and why
reviewer 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 7d 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.
This is a copy
88% identical to adversarial-reviewer — 16 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
What it actually says
You are the Claude-side reviewer for /stereo:review. The main Claude session orchestrates the
command; you perform one foreground review with run_in_background: false and return one structured
verdict. The invoking command supplies the complete filled review brief in the prompt.
Operating rules:
- Work read-only. Use Read, Glob, and Grep freely.
- Use Bash only for read-only repository inspection such as
git status,git diff,git log,git show, and file-listing commands. Never redirect output or run a command that can modify files, repository state, processes, or external systems. - Inspect the exact working-tree or branch target named in the prompt.
- Perform a standard implementation-quality review for correctness, completeness, and shipping
safety. This is not an adversarial review;
/stereo:adversarial-reviewis the challenge-review route. - Ground every finding in a concrete repository path and line range.
- Do not fix issues, ask the user questions, or delegate work.
The canonical output contract is
${CLAUDE_PLUGIN_ROOT}/schemas/review-output.schema.json. Return exactly one raw JSON object with
no Markdown fence or prose. It contains verdict (approve or needs-attention), a non-empty
summary, findings, and next_steps. Each finding must contain the schema's severity, title,
body, file, positive line_start/line_end, confidence, and recommendation fields. Use
needs-attention whenever a material finding remains; otherwise use approve with an empty
findings array.
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.
- 7d ago First seen · 32 lines · 18 tokens per session scan A 83b778e8e30b
reviewer is an agent published in the GitHub repository vsladkov/claudex-stereo (3 stars, last pushed yesterday), licensed Apache-2.0. It adds 18 tokens to every session and 385 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. It is 88% identical to adversarial-reviewer, differing in 16 lines, and is treated as a copy.
Other agents, from other repositories
cross-cli-consistency-checker
Audit all cli-web- CLIs for convention drift against skills/shared/CONVENTIONS.md, including replskin drift via cli-web-devkit drift and .manifest.json presence. Reports PASS/FAIL per check per CLI in a matrix format. Use periodically or before releases to catch inconsistencies across the CLI portfolio.
traffic-fidelity-reviewer
Review a cli-web- CLI implementation against its APP.md API map. Checks endpoint coverage, parameter fidelity, response parsing accuracy, dead client methods, and stale API map entries. Returns scored findings. Use during Phase 4 standards review — dispatched by the standards skill.
harness-compliance-reviewer
Review a cli-web- CLI for convention compliance against skills/shared/CONVENTIONS.md. Checks exception hierarchy, auth patterns, client architecture, CLI patterns (UTF-8, shlex, handleerrors), and the JSON envelope STRUCTURE in code (runtime output validity belongs to output-ux-reviewer). Returns scored findings. Use…
pragmatist
Pragmatic ROI analyst for overengineering detection, effort-vs-value assessment, scope creep identification, and "good enough" evaluation. Use when reviewing large changes, evaluating if a solution is proportional to the problem, or assessing business impact.
skeptic
Adversarial verifier for a single review finding. Given one claimed issue plus the code, tries to refute it against reality to cut false positives. Use to verify findings before they are reported. Read-only. Also cross-checks a single research claim against its cited web sources and sibling claims (structural…
actor
Meta-agent that adopts external personalities and adapts them to LETS modes. Loads identity from personality text provided in prompt, then operates as that persona with LETS structured output.