What is triage-rule-reviewer?

An analysis agent that looks for patterns in completed code-triage results that could become fixed classification rules. Code triage sorts possible entry points into categories such as real entry points, dead code, and false positives.

Use it to compare triage outcomes with function metadata and suggest patterns for deterministic rules based on exports, access levels, callback context, kinds, or file paths.

Fixed rules can classify recurring cases without another language-model investigation, which may reduce future triage work and delay.

How CRJFisher/ariadne configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to CRJFisher/ariadne. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/CRJFisher/ariadne/main/.claude/agents/triage-rule-reviewer.md

Clone the repo

git clone --depth 1 https://github.com/CRJFisher/ariadne

Made for: Claude Code.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/agents/crjfisher/ariadne/triage-rule-reviewer/github.svg)](https://agentmods.dev/agents/crjfisher/ariadne/triage-rule-reviewer)

Your own site

<a href="https://agentmods.dev/agents/crjfisher/ariadne/triage-rule-reviewer"><img src="https://agentmods.dev/badge/agents/crjfisher/ariadne/triage-rule-reviewer/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.

Your own site · 80×15

<a href="https://agentmods.dev/agents/crjfisher/ariadne/triage-rule-reviewer"><img src="https://agentmods.dev/badge/agents/crjfisher/ariadne/triage-rule-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 29 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 893 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00029	$0.00893
Opus 5	$0.00015	$0.00447
Sonnet 5	$0.00006	$0.00179
Haiku 4.5	$0.00003	$0.00089

Measured 11d ago against content hash 0348ff6bffd3, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

triage-rule-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 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.

.claude/agents/triage-rule-reviewer.md · 77 lines

How it starts

The opening of the file, as written. The whole thing — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Purpose

You analyze completed triage results to identify metadata patterns that could be encoded as deterministic classification rules. These rules would allow future triage runs to classify entries without LLM investigation, reducing cost and latency.

The triage pipeline classifies entry point candidates into three categories: true-positive (legitimate entry points — public API, framework hooks, CLI handlers), dead-code (unused/abandoned callables), and false-positive (callables with callers that Ariadne missed). Deterministic rules use entry metadata (export status, access modifier, kind, callback context, file path patterns) to classify entries without investigation. The classify_entrypoints function in .claude/skills/self-repair-pipeline/src/classify_entrypoints.ts implements existing rules.

Instructions

Read the triage state file at the path provided in your prompt. Parse the JSON to access the entries array with their results.
Read the analysis file referenced in analysis_file to access the full EnrichedFunctionEntry metadata for each entry.
Identify patterns across entries with the same classification. Look for correlations between:
- is_exported + classification outcome
- access_modifier + classification outcome
- kind (function/method/constructor) + classification outcome
- callback_context flags + classification outcome
- diagnostics.diagnosis value + classification outcome
- File path patterns (e.g., entries in test/ files, index.ts files)
- call_summary characteristics (e.g., zero total calls = dead code)
- group_id patterns from false-positive results
Read the existing deterministic rules in .claude/skills/self-repair-pipeline/src/classify_entrypoints.ts to avoid proposing rules that already exist.
Evaluate each candidate rule:
- How many entries would it correctly classify?
- Are there any entries it would misclassify?
- Assign a confidence rating:
  - HIGH: 100% accuracy on training data, clear logical basis, safe to auto-apply
  - MEDIUM: 90%+ accuracy, reasonable basis, should be human-reviewed before adding
  - LOW: Interesting pattern but insufficient data or edge cases exist, informational only

Read the full file on GitHub · 77 lines

Changes

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 · 77 lines · 29 tokens per session scan A 0348ff6bffd3

Subscribe to this mod's changes

triage-rule-reviewer is an agent published in the GitHub repository CRJFisher/ariadne (22 stars, last pushed 8d ago), licensed MIT. It adds 29 tokens to every session and 893 once invoked, about $0.0001 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 2026-08-30.

Code review Data and AI ast code-intelligence nodejs parser tree-sitter

rag-pipeline-reviewer

Reviews RAG (Retrieval-Augmented Generation) pipelines for retrieval quality, chunking strategy, embedding choices, and evaluation coverage. Invoke when the user builds, modifies, or debugs a RAG system, vector store integration, or asks about retrieval accuracy.

affaan-m/ECC · 58 tokens

ag2-reviewer

Reviews AG2 agent code for tool contract violations, prompt quality, security issues, and best practices. Invoke after creating or modifying AG2 agents.

davepoon/buildwithclaude · 34 tokens

geo-routing-engineer

Geospatial and routing specialist for Product-Builder products with maps, scheduling-by-location, or vehicle routing (route-optimization in logistics, dispatch in home services, field-booking). Owns the routing contract — geocoding, the VRP/routing model (constraints, objective), maps/distance-matrix provider…

avelikiy/great_cto · 112 tokens

ariadne: Agent for Claude Code