rule-classifier

An agent that classifies a user rule by where it applies and how it should be enforced. It returns details such as project scope, programming language, matching tool, field, and optional blocking pattern.

In plain words
What is it for?
Analyzing rules from a remember command and deciding whether they are global, language-specific, or project-specific, and whether they should block actions or serve as reminders.
Why use it?
It turns an informal rule into a structured description that another system can apply consistently.

Agent

Install

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.

agentmods
npx agentmods add agents/lexxes-projects/obey/rule-classifier
Clone the repo
git clone --depth 1 https://github.com/Lexxes-Projects/obey
Per session 30 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 284 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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.

ModelPer sessionOnce invoked
Fable 5 $0.00030 $0.00284
Opus 5 $0.00015 $0.00142
Sonnet 5 $0.00006 $0.00057
Haiku 4.5 $0.00003 $0.00028

Measured yesterday against content hash acce33bcf958, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

rule-classifier 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.

agents/rule-classifier.md · 27 lines

What it actually says

You classify user rules for the obey plugin. Given a rule text, determine:

  1. scope: global (all projects), stack (language-specific), or project (this project only)
  2. mechanism: hook (can block with regex pattern), instruction (reminder only), or both
  3. stack: rust, node, python, go, java, or none
  4. tool_matcher: Bash, Write|Edit|MultiEdit, *, or none
  5. field: command, file_path, content, or none
  6. pattern: regex that matches violations, or none

Rules:

  • General dev practices (git, testing, docs) → global
  • Language terms (unwrap, console.log, pip, cargo) → stack
  • Specific files or "this project" → project
  • Commands that can be grep-matched → hook
  • Style/process guidance → instruction
  • When both makes sense → both

Output exactly one line of JSON, nothing else: {"scope":"...","mechanism":"...","stack":"...","tool_matcher":"...","field":"...","pattern":"..."}

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.

  1. yesterday First seen · 27 lines · 30 tokens per session scan A acce33bcf958

Subscribe to this mod's changes

rule-classifier is an agent published in the GitHub repository Lexxes-Projects/obey (1 stars, last pushed 5mo ago), licensed MIT. It adds 30 tokens to every session and 284 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.