agentmods.dev

CluxMate: Instructions file for Claude Code

AGENTS.md

CluxMate AGENTS.md is an instructions file for Claude Code, Codex, OpenCode from r1c7/CluxMate. It costs 15,399 tokens per session, scanned B, original, MIT.

A project guide for CluxMate, a Python coding agent with command-line, text-based, and desktop interfaces.

In plain words
What is it for?
Use it when working on CluxMate code, especially changes involving its agent core, user interfaces, permissions, sandbox, or hooks.
Why use it?
It gives the agent shared knowledge about the project’s architecture, security rules, risk modes, and repository setup.

Instructions file for Claude CodeCodexOpenCode

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

This is r1c7/CluxMate's own configuration. It tells Claude Code, Codex and OpenCode how to work on CluxMate 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 CluxMate configures →

not rated 140repo +1 changed 1mo ago B scan 15,399 tokens original MIT
View source ↗ r1c7/CluxMate
Reuse

Borrowing it

Nothing to install: this file belongs to r1c7/CluxMate. 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/r1c7/CluxMate/master/AGENTS.md
Clone the repo
git clone --depth 1 https://github.com/r1c7/CluxMate

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 CluxMate AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/r1c7/cluxmate/agents-md/github.svg)](https://agentmods.dev/instructions/r1c7/cluxmate/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/r1c7/cluxmate/agents-md"><img src="https://agentmods.dev/badge/instructions/r1c7/cluxmate/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 CluxMate AGENTS.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/r1c7/cluxmate/agents-md"><img src="https://agentmods.dev/badge/instructions/r1c7/cluxmate/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 15,399 This file is loaded in full into every session.
When invoked 15,399 The same file — it is already loaded in full.
Security scan B 3 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.15399 $0.15399
Opus 5.5 $0.06160 $0.06160
Sonnet 5.5 $0.03080 $0.03080
Haiku 4.5 $0.01540 $0.01540

Measured 1mo ago against content hash 215121d4eefe, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade B, and why

CluxMate AGENTS.md scanned grade B with 3 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 1mo 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.

Asks for rootlowPrivilege escalation

A mod that escalates privileges can change anything on the machine, not only the project.

- Risk levels `safe` < `write` < `dangerous` < `critical`. `critical` (bash `format`/`mkfs`/`dd`/`> /dev/`/`chmod 777`) never auto-approves **where a policy is consulted (TUI/desktop — headless has no gate, see above)**

Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.

Reaches for credential filesmediumPrivilege escalation

SSH keys, cloud credentials, git-credentials, .npmrc, /etc/shadow: reading these is how a config file becomes a credential leak.

- **Read denylist** — `~/.cluxmate/forbid-read.json` (`core/read_denies.py`, user-global): an explicit `paths` list plus an opt-in `protect_sensitive` toggle (default OFF = zero behavior change) adding `.env` / `.git-cre

Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.

Runs shell commandslowCapability

Expected in a hook, worth knowing in a rule or an instructions file.

- **Project root ≠ session cwd** (`core/project_root.py` — the ONE place this policy lives): `resolve(cwd).config_root` is the session cwd EXCEPT inside a LINKED git worktree (git-dir ≠ git-common-dir), where it is the m
AGENTS.md · 175 lines

How it starts

The opening of the file, as written. The whole thing — 175 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Project memory

What CluxMate is

  • CluxMate = one Python agent core (cluxmate/) driving three front-ends: headless CLI (cluxmate -p "..."), Textual TUI (cluxmate), and an Electron desktop (desktop/) over JSON-RPC stdio (cluxmate agent stdio). Providers are OpenAI-compatible only (core/providers/ = base + openai + factory; no Anthropic provider).
  • Four risk modes (plan / default / acceptEdits / yolo), per-session only, never persisted: plan registers ONLY read tools (no bash/write/delete/task/update_memory/MCP — hard isolation, nothing to approve); default auto-approves safe and asks for write/dangerous; acceptEdits auto-approves write; yolo disarms the sandbox boundaries entirely (the one explicit opt-out). Approval is a FRONT-END feature, not an agent-loop one: PermissionPolicy is instantiated only in the JSON-RPC server and the TUI, and the loop itself never consults it — AgentCallbacks.on_tool_start defaults to ToolDecision(True, "auto"), so headless cluxmate -p auto-approves every call (including critical and escalation) and relies on the two enforcement boundaries instead of prompts. Any claim that "risky tools always prompt" holds for TUI/desktop only.
  • Risk levels safe < write < dangerous < critical. critical (bash format/mkfs/dd/> /dev//chmod 777) never auto-approves where a policy is consulted (TUI/desktop — headless has no gate, see above) and is never always-allowable, like escalation. Running code (interpreters/build tools/scripts — python, node, npm run, ./x.sh, …) is dangerous, never safe; classified by runner name with a run fallback for unlisted executables.
  • Always-allow lives in permissions.json in two tiers: always_allow_tools (write risk) and always_allow_dangerous_tools (dangerous risk — delete_file whole-tool, and bash PER category "bash:rm" / "bash:git-reset-hard" / "bash:python" / "bash:run" / …, NEVER a bare "bash"). Escalation (danger-full-access), critical bash and MCP dangerous tools are never always-allowable.
  • The repo root IS the git repo (branch master, remote [email protected]:r1c7/CluxMate.git). .gitignore excludes docs/, scripts/, .cluxmate/; snapshots/*.png (UI screenshots) ARE tracked. Undo checkpoints are a SEPARATE shadow repo at ~/.cluxmate/checkpoints/<sha1(cwd)>.git (env GIT_DIR/GIT_WORK_TREE + GIT_CONFIG_GLOBAL=/dev/null) and never touch the user's repo — core/checkpoints.py.

Read the full file on GitHub · 175 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. 1mo ago Changed · +302 tokens per session 215121d4eefe
  2. 1mo ago Changed · +3 lines · +1,131 tokens per session 35413df73522
  3. 1mo ago Changed · +24 lines · +3,505 tokens per session d5f1e694b967
  4. 1mo ago Changed · +10 lines · +1,038 tokens per session 7a4b3ee640ae
  5. 1mo ago Changed · -14 lines · -5,034 tokens per session 0f753e2f6e51
  6. 1mo ago Changed · +1 lines · +848 tokens per session 095262e12eb0
  7. 1mo ago Changed · +9 lines · +1,049 tokens per session 48f3ff5c6a65
  8. 1mo ago Changed · +29 lines · +3,155 tokens per session 28adf90131ea
  9. 1mo ago Changed · +8 lines · +867 tokens per session 55af4d60898f
  10. 2mo ago First seen · 105 lines · 8,538 tokens per session scan B 04aec69df836

Subscribe to this mod's changes

CluxMate AGENTS.md is an instructions file published in the GitHub repository r1c7/CluxMate (140 stars, last pushed 1mo ago), licensed MIT. It adds 15,399 tokens to every session, about $0.0616 per session on Opus 5.5. A static security scan graded it B with 3 findings (asks for root, reaches for credential files, runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

agentloopai-agentcode-assistantcoding-agentevent-sourcingsandbox
Related

CluxMate AGENTS.md compared with instructions from other repositories

CluxMate AGENTS.md next to 6 instructions from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
CluxMate AGENTS.mdr1c7/CluxMateInstructions file1401mo agoMITB15,399Claude Code, Codex, OpenCode
vscode buildNext.instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA9,719GitHub Copilot
next.js AGENTS.mdvercel/next.jsInstructions file143ktodayMITA7,461Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchainInstructions file149kyesterdayMITA4,680Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA5,516GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA3,093GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harnessInstructions file274k2d agoMITA4,313Codex, OpenCode