agentmods.dev

Plexus CLAUDE.md

Detailed project guidance for Plexus, a local web dashboard for managing AI-agent settings on one computer. It brings together tool-server connections, reusable skill bundles, and instruction files for several coding agents.

In plain words
What is it for?
Use it when developing or validating Plexus features that manage Claude Code, Cursor, Codex, Gemini CLI, Qwen Code, Factory Droid, DeepSeek Harness, or OpenCode settings.
Why use it?
It gives agents a shared map of the product, supported configuration files, repository structure, and important rules before they change code.

Instructions file

28 4d ago C scan 9,083 tokens original Apache-2.0
View the source file miniLV/Plexus
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 instructions/minilv/plexus/claude-md
Clone the repo
git clone --depth 1 https://github.com/miniLV/Plexus
Per session 9,083 This file is loaded in full into every session.
When invoked 9,083 The same file — it is already loaded in full.
Security scan C 3 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 $0.09083 $0.09083
Opus 5 $0.04542 $0.04542
Sonnet 5 $0.01817 $0.01817
Haiku 4.5 $0.00908 $0.00908

Measured 2d ago against content hash 34319658fb5f, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade C, and why

Plexus CLAUDE.md scanned grade C 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 2d 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.

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

| Codex | `~/.codex/config.toml` | `~/.codex/skills/` | `shared` |

Enumerates other installed skillsmediumAgent snooping

Other skills' SKILL.md files reveal prompts, capabilities and secrets that should be invisible to peers.

ls ~/.claude/skills | grep plexus-backup || echo "(clean)"

Makes network callslowCapability

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

curl -s -o /dev/null -w "%{http_code}\n" http://localhost:7777/
CLAUDE.md · 1,044 lines

How it starts

The opening of the file, as written. The whole thing — 1,044 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md - Plexus

For the next Claude Code / Droid / Codex session. This is the working map for Plexus: what the product is, how the repo is shaped today, which invariants are load-bearing, and how to validate changes. Read it before touching code.

1. What Plexus Is

Plexus is a local-only web dashboard for managing AI agent configuration on one machine. It centralizes:

  • MCP servers
  • Skills / command bundles
  • User-level instruction files such as CLAUDE.md and AGENTS.md

Supported built-in agents:

Agent MCP file Skill / prompt dir MCP mode
Claude Code ~/.claude.json ~/.claude/skills/ shared
Cursor ~/.cursor/mcp.json ~/.cursor/commands/ exclusive
Codex ~/.codex/config.toml ~/.codex/skills/ shared
Gemini CLI ~/.gemini/settings.json ~/.gemini/skills/ shared
Qwen Code ~/.qwen/settings.json ~/.qwen/skills/ shared
Factory Droid ~/.factory/mcp.json ~/.factory/skills/ exclusive
DeepSeek Harness ~/.dsh/mcp.json ~/.dsh/skills/ shared
OpenCode ~/.config/opencode/opencode.json ~/.config/opencode/skills/ shared

The current package version is tracked in the root package.json. All workspace package.json files must stay on the same version:

  • package.json
  • apps/web/package.json
  • packages/core/package.json
  • packages/cli/package.json

Use Git tags / releases and the workspace package.json files as the version source of truth. Do not treat this file as a release ledger.

Product Contract

Plexus keeps a canonical store under ~/.config/plexus/ and projects it into agent-native locations with the smallest reversible write:

  • Exclusive MCP files are rendered into ~/.config/plexus/.cache/mcp/<agent>.json, then linked or copied into the agent path.
  • Shared MCP files are partial-written in place; Plexus rewrites only the MCP section and preserves unrelated auth/history/profile keys.
  • Skills are linked or copied from the Plexus store into each agent's native skill directory.
  • Rules are stored once at ~/.config/plexus/personal/rules/global.md, then linked or copied to each built-in agent's native instruction file (CLAUDE.md, AGENTS.md, GEMINI.md, or QWEN.md).
  • Backups are taken before syncs, toggles, and dashboard file edits.

Read the full file on GitHub · 1,044 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. 2d ago First seen · 1,044 lines · 9,083 tokens per session scan C 34319658fb5f

Subscribe to this mod's changes

Plexus CLAUDE.md is an instructions file published in the GitHub repository miniLV/Plexus (28 stars, last pushed 4d ago), licensed Apache-2.0. It adds 9,083 tokens to every session, about $0.0454 per session on Opus 5. A static security scan graded it C with 3 findings (reads agent configuration directories, enumerates other installed skills, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Analytics and metricsagent-configagent-configurationai-agentsai-coding-toolsclaude-codecodex
Related

Other instructions, from other repositories

aider-desk AGENTS.md

AGENTS.md instructions for hotovo/aider-desk, covering agents.md, common commands, development, type checking and linting and formatting.

hotovo/aider-desk · 1,725 tokens

ai-setup AGENTS.md

Instructions for caliber-ai-org/ai-setup, covering caliber, commands, architecture, conventions and before committing.

caliber-ai-org/ai-setup · 1,076 tokens

PPT-Design-Skill CLAUDE.md

Instructions for sunchaokun/PPT-Design-Skill: Use the ppt-design-skill workflow for presentation design tasks. The published pptx-designer Python library creates the editable PPTX; this repository owns the design process, installation, rendering, and PNG review.

sunchaokun/PPT-Design-Skill · 139 tokens

githits-cli AGENTS.md

Instructions for githits-com/githits-cli, covering githits agent instructions, general, architecture, testing and development workflow (docs-driven).

githits-com/githits-cli · 3,113 tokens

claude-code-codex-bridge AGENTS.md

Instructions for vladolaru/claude-code-codex-bridge, covering agents.md, repository overview, architecture, repository structure and package layout.

vladolaru/claude-code-codex-bridge · 2,837 tokens

kodus-ai AGENTS.md

AGENTS.md instructions for kodustech/kodus-ai, covering kodus ai, structure, stack, getting started and commands.

kodustech/kodus-ai · 1,004 tokens