claude-model-guard: Instructions file for Claude Code

AGENTS.md

claude-model-guard AGENTS.md is an instructions file for Claude Code, Codex, OpenCode from ventusff/claude-model-guard. It costs 2,137 tokens per session, scanned A, original, MIT.

A set of instructions and scripts for a Claude Code plugin that shows when the model has been silently downgraded. It can stop a turn and enter model and effort commands in the session's terminal to undo a safeguard flag.

In plain words
What is it for?
Use it to inspect the plugin's files, hooks, status display, installation, and verification rules when working on the model-guard project.
Why use it?
It helps expose unexpected model changes instead of letting a task continue with a weaker or different model without notice.

Instructions file for Claude CodeCodexOpenCode

Written for Claude Code and Codex and OpenCode: SessionStart hook event, but also the file is AGENTS.md. Also seen: reads .claude/ paths; mentions Claude Code; mentions AGENTS.md.

This is ventusff/claude-model-guard's own configuration. It tells Claude Code, Codex and OpenCode how to work on claude-model-guard itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything claude-model-guard configures →

Reuse

Borrowing it

Nothing to install: this file belongs to ventusff/claude-model-guard. 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/ventusff/claude-model-guard/main/AGENTS.md
Clone the repo
git clone --depth 1 https://github.com/ventusff/claude-model-guard

Made for: Claude Code, Codex, OpenCode.

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.

agentmods badge for claude-model-guard AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/ventusff/claude-model-guard/agents-md/github.svg)](https://agentmods.dev/instructions/ventusff/claude-model-guard/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/ventusff/claude-model-guard/agents-md"><img src="https://agentmods.dev/badge/instructions/ventusff/claude-model-guard/agents-md/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.

agentmods 80×15 button for claude-model-guard AGENTS.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/ventusff/claude-model-guard/agents-md"><img src="https://agentmods.dev/badge/instructions/ventusff/claude-model-guard/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 2,137 This file is loaded in full into every session.
When invoked 2,137 The same file — it is already loaded in full.
Security scan A 1 finding. 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.02137 $0.02137
Opus 5 $0.01069 $0.01069
Sonnet 5 $0.00427 $0.00427
Haiku 4.5 $0.00214 $0.00214

Measured 3d ago against content hash 2b5d8f17ae5b, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-13, from the pricing page.

Security

Grade A, and why

claude-model-guard AGENTS.md scanned grade A with 1 finding 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 3d 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

Requirements: bash 4+, jq and curl; the recovery path also uses `flock` and `setsid`. There is no CI workflow — `tests/run.sh` on the machine is the bar. Hooks load at session start, so a change under `hooks/` or `script
AGENTS.md · 55 lines

How it starts

The opening of the file, as written. The whole thing — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md — claude-model-guard

A Claude Code plugin (model-guard, distributed from this repo as marketplace claude-model-guard): a full-width statusline band that flags a silent model downgrade, plus hooks that undo a safeguard-flag downgrade — stop the turn, type /model and /effort into the session's own terminal (tmux, zellij or kitty), resume the task. Pure bash + jq, no build step, no daemon. Installed via /plugin marketplace add ventusff/claude-model-guard, so main on GitHub is what every install fetches.

The Codex integration lives independently under codex/model-guard/: a valid Codex plugin and Python package, with a pinned native Codex build with metadata events and conditional footer warnings; the Python app-server adapter is used only by the explicit standalone probe. See its ROUTING.md for evidence boundaries and primary sources. It monitors disclosed routing and does not perform automatic model recovery.

Layout

Path What
.claude-plugin/plugin.json Plugin manifest: name, version, description, keywords.
.claude-plugin/marketplace.json The marketplace this repo is; lists the one plugin with source: ./.
hooks/hooks.json Registers scripts/guard-hook.sh on SessionStart, SessionEnd, PreModelSwitch, PostModelSwitch, PreToolUse, Stop, UserPromptSubmit; check-install.sh on SessionStart.
scripts/statusline.sh The statusline. Sources lib.sh next to it. setup copies the three scripts to ~/.claude/model-guard/; that copy is what Claude Code runs. Asks Claude Code's usage endpoint for the logged-in account's rate-limit usage, cached in usage.json under the state dir.
scripts/lib.sh Shared helpers (mg_*): config, language, model strength ranking, saved effort, per-session state, keystroke channels. Sourced by the statusline, the hook and the driver. Holds MG_VERSION.
scripts/text.sh Every user-facing string in the eight band languages (mg_text). Sourced by lib.sh.
scripts/guard-hook.sh The one hook command; dispatches on hook_event_name, answers with hook JSON.
scripts/recover.sh The detached driver that types Esc, /model, /effort, continue prompt into the terminal.
scripts/check-install.sh SessionStart: hints until setup is done; refreshes the installed scripts when the plugin ships a newer MG_VERSION; turns a registered flat ~/.claude/model-guard.sh into a hand-off to the directory.
skills/setup/SKILL.md, skills/remove/SKILL.md /model-guard:setup and /model-guard:remove — step lists the agent follows, not shell scripts.
tests/run.sh The test suite: synthetic hook payloads, dryrun channel, fake tmux/zellij binaries, isolated state, config and install dir.
README.md, README.zh-CN.md, CHANGELOG.md User docs in both languages, and the release log.
assets/ The README's hero and setup SVGs.
codex/model-guard/model_guard/ Codex probe metadata (state.py, reasoning.py, check.py, relay.py), running-session discovery (sessions.py), native installer and diagnostics.
codex/model-guard/tests/ Python unit tests and opt-in real Codex/HTTP/WebSocket/PTY fixtures.

Read the full file on GitHub · 55 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.

  1. 3d ago Changed · +14 lines · +889 tokens per session 2b5d8f17ae5b
  2. 4d ago Changed · +70 tokens per session 8473c0d9b42d
  3. 6d ago First seen · 41 lines · 1,178 tokens per session scan A 19d752191c2b

Subscribe to this mod's changes

claude-model-guard AGENTS.md is an instructions file published in the GitHub repository ventusff/claude-model-guard (2 stars, last pushed 3d ago), licensed MIT. It adds 2,137 tokens to every session, about $0.0107 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-07.

Related

Other instructions, from other repositories

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

next.js AGENTS.md

AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,153 tokens

vscode oss-third-party-notices.instructions.md

Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).

microsoft/vscode · 5,001 tokens

spec-kit AGENTS.md

AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.

github/spec-kit · 7,126 tokens

langchain AGENTS.md

AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.

langchain-ai/langchain · 4,469 tokens