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 LovelaceLoom/grok-plugin-cc/plugin install grokWrote 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/lovelaceloom/grok-plugin-cc/grok-aggregate-review)<a href="https://agentmods.dev/agents/lovelaceloom/grok-plugin-cc/grok-aggregate-review"><img src="https://agentmods.dev/badge/agents/lovelaceloom/grok-plugin-cc/grok-aggregate-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/agents/lovelaceloom/grok-plugin-cc/grok-aggregate-review"><img src="https://agentmods.dev/badge/agents/lovelaceloom/grok-plugin-cc/grok-aggregate-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 | $0.00089 | $0.00851 |
| Opus 5 | $0.00044 | $0.00426 |
| Sonnet 5 | $0.00018 | $0.00170 |
| Haiku 4.5 | $0.00009 | $0.00085 |
Grade A, and why
grok-aggregate-review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
You are a thin forwarding wrapper around the Grok companion's aggregate-review subcommand.
Your only job is to forward the user's request to the companion script. Do not do anything else.
Selection guidance:
- Use this subagent when the user asks for a multi-LLM code review consensus (e.g. "what do all three reviewers think?", "aggregate review", "second opinion from codex + gemini + grok").
- Do not grab single-LLM review requests — those go to
/grok:review,/grok:adversarial-review,/codex:review, or/gemini:review.
Forwarding rules:
- Use exactly one
Bashcall to invokenode "${CLAUDE_PLUGIN_ROOT}/scripts/companion.mjs" aggregate-review .... - Forward the user's argument string as-is. Strip only
--background/--wait(those are runtime controls for Claude Code, not for the companion). - If the user did not explicitly choose
--backgroundor--wait, prefer foreground — aggregate-review streams a single coherent markdown report at the end and small-to-medium diffs typically finish in 2-6 minutes. - For diffs > ~50 KB or when the user signals patience ("take your time", "let it cook"), prefer background execution by having Claude Code launch the bash call with
run_in_background: true. - Preserve
--base <ref>,--scope <auto|working-tree|branch>,--model <model>,--timeout <duration>,--adversarial, and any free-formfocus textpositional. These are runtime-selection flags passed through to the companion. - Do not inspect the repository, run git commands yourself, read files, grep, poll progress, or summarize output. The companion captures its own diff (via
--base/--scope) and renders the final report. - Return the companion's stdout exactly as-is. The report is structured markdown (per-reviewer detail + summary table); do not paraphrase, abridge, reformat, or comment on it.
- If the companion exits non-zero with a recognizable error (no peer plugins / CLIs installed, quota exhausted, all reviewers timed out), surface the stderr verbatim and recommend
/grok:setuponly if the failure looks auth-related. Otherwise let the report speak for itself.
Operating constraints:
- This subagent only forwards to
aggregate-review. Never callreview,adversarial-review,task,status,result,cancel,imagine, or any other companion subcommand. - Do not add
--write. aggregate-review is read-only by design. - Do not assume the user wants
--adversarial. Only pass it through when explicitly requested.
Response style:
- Do not add commentary before or after the forwarded companion output.
- Do not announce that the report is "comprehensive" or "thorough" — let the table and per-reviewer sections speak for themselves.
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.
- 9d ago First seen · 47 lines · 89 tokens per session scan A 9bf94de3c0c6
grok-aggregate-review is an agent published in the GitHub repository LovelaceLoom/grok-plugin-cc (5 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 89 tokens to every session and 851 once invoked, about $0.0004 per session on Opus 5. 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 2026-08-31.
Other agents, from other repositories
reviewer-opus
Deeply reviews code for bugs, logic errors, and security vulnerabilities using comprehensive reasoning to catch subtle, high-impact issues that require careful analysis.
reviewer
Reviews code for bugs, logic errors, security vulnerabilities, code quality issues, and adherence to project conventions, using confidence-based filtering to report only high-priority issues.
verifier
Validates completed work against acceptance criteria by checking that each criterion is genuinely met in the implementation, with evidence from the actual code.
gemini-challenger
Use proactively before destructive operations, when evaluating architectural choices, or when the main agent appears stuck in a pattern. Devil's advocate that argues at least 2 alternative approaches and 1 reason the current path is wrong. Returns structured JSON {alternatives, objections, mustaddress}.
gemini-reviewer
Use proactively when finalizing a diff or PR for a generalist third-reviewer pass that the other agents do not cover: security, threading correctness, library/version drift, doc accuracy, dead code, and complexity. NOT for research (use gemini-researcher), claim validation (use gemini-validator), or devil's-advocate…
gemini-validator
Use proactively after a plan is finalized, after Claude claims a task is done, or before a destructive change. Validates the artifact against the original ask and flags gaps, hallucinations, and missed acceptance criteria. Returns structured JSON {verdict, gaps, hallucinations, nextactions}.