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 instructions/r1c7/cluxmate/agents-mdgit clone --depth 1 https://github.com/r1c7/CluxMateWrote 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.
[](https://agentmods.dev/instructions/r1c7/cluxmate/agents-md)<a href="https://agentmods.dev/instructions/r1c7/cluxmate/agents-md"><img src="https://agentmods.dev/badge/instructions/r1c7/cluxmate/agents-md.svg" alt="Measured on agentmods" height="20"></a>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 | $0.07236 | $0.07236 |
| Opus 5 | $0.03618 | $0.03618 |
| Sonnet 5 | $0.01447 | $0.01447 |
| Haiku 4.5 | $0.00724 | $0.00724 |
Grade B, and why
CluxMate AGENTS.md scanned grade B with 2 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 today.
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.
- Four risk modes (`plan` / `default` / `acceptEdits` / `yolo`): `plan` registers only read tools (no bash/write/delete/task/update_memory/MCP — hard isolation, nothing to approve); `default` approves `safe` automaticall 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 + built-in sensitive-file template** (shipped 2026-09-03, the P0 "敏感文件读保护默认值"): `~/.cluxmate/forbid-read.json` = `{"protect_sensitive": bool, "paths": [...]}` (`core/read_denies.py`, user-global; legacy Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.
How it starts
The opening of the file, as written. The whole thing — 87 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/) that drives the core 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):planregisters only read tools (no bash/write/delete/task/update_memory/MCP — hard isolation, nothing to approve);defaultapprovessafeautomatically, asks forwrite/dangerous;acceptEditsauto-approveswrite;yolodisarms the sandbox boundaries entirely (the one explicit opt-out). Mode is per-session only, never persisted. Risk levels aresafe<write<dangerous<critical;critical(bashformat/mkfs/dd/> /dev//chmod 777) never auto-approves and is never always-allowable, like escalation. Running code (interpreters/build tools/scripts —python,node,npm run,./x.sh, …) isdangerous(neversafe), classified by runner name with arunfallback for unlisted executables. Permissions.json stores two always-allow tiers:always_allow_tools(write risk) andalways_allow_dangerous_tools(dangerous risk —delete_filewhole-tool, andbashPER category"bash:rm"/"bash:git-reset-hard"/"bash:python"/"bash:run"/…, never a bare"bash"; escalationdanger-full-access,criticalbash, and MCP dangerous tools are never always-allowable). - The repo root IS a git repo (
.git/, branchmaster, remote[email protected]:r1c7/CluxMate.git, init 2026-08-25, actively committed since)..gitignoreexcludesdocs/,scripts/,.cluxmate/;snapshots/*.png(UI screenshots) ARE tracked. Checkpoint shadow-git repos live separately in~/.cluxmate/checkpoints/<sha1(cwd)>.git(seecore/checkpoints.py) and never touch the user's real repo (GIT_DIR/GIT_WORK_TREE set via env +GIT_CONFIG_GLOBAL=/dev/null).
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.
- today Changed · +2 lines · +1,989 tokens per session scan A → B e5ae5cc26eed
- 4d ago First seen · 85 lines · 5,247 tokens per session scan A 978c135e325c
CluxMate AGENTS.md is an instructions file published in the GitHub repository r1c7/CluxMate (44 stars, last pushed today), licensed MIT. It adds 7,236 tokens to every session, about $0.0362 per session on Opus 5. A static security scan graded it B with 2 findings (asks for root, reaches for credential files). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
zero AGENTS.md
AGENTS.md instructions for Gitlawb/zero, covering repository guidelines for zero, 1. contribution and pull request rules, 2. repository and implementation conventions, 3. required validation and 4. common review blockers.
memU AGENTS.md
AGENTS.md instructions for NevaMind-AI/memU, covering agents.md, mission, layer map, implementation rules and backend parity.
LaPis AGENTS.md
Instructions for GeneGulanesJr/LaPis, covering protocols, 1. code & doc retrieval — use memory-code / memory-doc, 2. persistent memory — automatic, when to use the tools and content format.
agent-smith CLAUDE.md
Claude Code instructions for holgerleichsenring/agent-smith, covering agent-smith — ai agent instructions, context files (read in this order), phase directory structure, experiential memory (remember / recall) and minting a phase id.
Core-Memory CLAUDE.md
Instructions for JohnnyFiv3r/Core-Memory, covering claude.md — core memory, what this repo is, guiding principle — engineering simplicity, boring primitives, rich views and mapping to the current codebase.
CodeAuto AGENTS.md
Instructions for Yulin-Bi/CodeAuto, a project described as: CodeAuto is an intelligent programming assistant developed using Java, drawing inspiration from Claude Code and Minicode.