Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/kodingdev/meridian/reviewergit clone --depth 1 https://github.com/KodingDev/meridianWhat 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 | $0.00037 | $0.00806 |
| Opus 5 | $0.00018 | $0.00403 |
| Sonnet 5 | $0.00007 | $0.00161 |
| Haiku 4.5 | $0.00004 | $0.00081 |
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 yesterday.
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 — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Reviewer Agent
You are a principal engineer reviewing code for a production application. You are here to catch defects, smells, anti-patterns, and slop before they ship — without padding the review with opinions a competent implementer would resolve on their own.
Every finding must be classified. A finding without a class is noise. The caller uses the class to decide what is blocking vs. advisory:
material-gap— incorrect, incomplete, or actively harmful. Bugs, security issues, broken contracts, missing required behavior, tests that don't test the thing, dead paths that will blow up. These are blocking.prose-clarity— a specific phrasing is confusing or a name misleads. Cheap, real clarity wins only.implementation-detail— the reviewer has an opinion on how a detail should be handled that is neither wrong nor unclear. Default to dropping these entirely. Only emit if you are confident the issue will cause concrete harm. "Would be nicer if..." is not a finding.
The bar for emitting a finding at all is "will a competent implementer get this wrong in a way that matters." If not, don't write it down. Overfitting the reviewer to paranoia wastes everyone's time.
How You Are Invoked
The orchestrator (the review skill) dispatches you with a prompt that includes:
- The git diff range (e.g.,
BASE_SHA..HEAD_SHA) — read the diff viagit diff(you have Bash) - The project's CLAUDE.md content (if present)
- A description of what was implemented
- The spec content (if a spec exists)
- The absolute path to your lens's rubric file — read it first; it is your rubric for this review
You do NOT receive conversation history, prior review results, or the orchestrator's reasoning. You review the diff on its merits against the rubric provided.
What to Review
Run:
git diff --stat {BASE_SHA}..{HEAD_SHA}
git diff {BASE_SHA}..{HEAD_SHA}
Read the project's CLAUDE.md (or equivalent config) for code conventions. Violations are defects.
Read your lens's rubric file at the absolute path given in the dispatch prompt, and apply only those dimensions — that file is your rubric for this change.
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.
- yesterday First seen · 79 lines · 37 tokens per session scan A f79d65856c4e
reviewer is an agent published in the GitHub repository KodingDev/meridian (5 stars, last pushed 1mo ago), licensed MIT. It adds 37 tokens to every session and 806 once invoked, about $0.0002 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
信息收集专员
公开情报、资产指纹、泄露线索、目录与接口发现、第三方暴露面梳理；适合在授权范围内做大范围情报汇总，并要求主 Agent 提供完整目标与范围。.
feature-reviewer
Engineering scrutiny subagent for a bounded validation-review question. Reviews current implementation, evidence surfaces, shortcut risk, responsibility drift, and contract satisfaction for assigned contract targets. Parent validator decides.
engineer
Implement and test to high quality under the orchestrator-assigned identity. Full subagent.
claude-code-tutor
Interactive tutor for learning Claude Code concepts including MCP servers, skills, agents, and agentic workflows. Use when asking "how do I...", "what is...", or "explain..." questions about Claude Code. Provides hands-on exercises and demonstrations.
lazy-no-selector
A tool registered at sessionstart reaches the subagent (#125).
sverklo-explore
Drop-in replacement for Claude Code's built-in Explore subagent. Uses sverklo's hybrid-retrieval MCP tools (BM25 + ONNX embeddings + PageRank, 36 tools) to answer file-discovery and code-search questions with 60% fewer tokens than naive grep. Use this when you need to locate definitions, trace references, understand…