Agent Plugin
Portable Agent Plugins package with skills and a gated MCP for ops and coding agents
tokenmaxxing ships a portable Agent Plugins 1.0 package at agent-plugin/. Compatible clients load that directory as a local plugin. Skills under skills/. MCP from mcp.json. Cursor, Codex, ChatGPT, VS Code, Copilot, and friends.
Install
Point the client at agent-plugin/ in a git checkout or an installed npm package (node_modules/tokenmaxxing/agent-plugin).
- Cursor: load
./agent-pluginas an Agent Plugin, or use the project.cursor/mcp.jsonstdio entry that runsbun ./agent-plugin/bin/tokenmaxxing-mcp. - Pi: project
.pi/settings.jsonalready listsagentPluginPaths: ["./agent-plugin"]. - Other Agent Plugins clients: add the same directory as a local plugin path.
The MCP binary is agent-plugin/bin/tokenmaxxing-mcp (Bun). It refuses ambient CLAUDE_CONFIG_DIR / CLAUDE_SECURESTORAGE_CONFIG_DIR. Set TOKENMAXXING_HOME for hermetic use.
MCP tools
Always available (read / free):
| Tool | Backing |
|---|---|
pool_ls | tokenmaxxing ls |
pool_status | tokenmaxxing status (never --force) |
doctor | tokenmaxxing doctor |
config_get | tokenmaxxing config / config get |
help | tool catalog + safety rules |
Mutating (dual-gated): each call needs confirm: true and the server env TOKENMAXXING_AGENT_MUTATIONS=1. Set that only after the user approves pool mutations.
| Tool | Backing |
|---|---|
pool_switch | Claude switch or Codex switch --codex when codex=true |
pool_check | tokenmaxxing check |
config_set / config_unset | config writers |
Not exposed: status --force, credential dumps, init / add / auth / rm / uninstall, process kills.
Tool results scrub bearer-looking and token-shaped spans. Prefer MCP over raw shell for pool ops.
Skills
| Skill | Job |
|---|---|
pool-status | Safe pool reads |
doctor-diagnostics | Install health |
switching-policy | GREEDY vs HARD, pace pressure, Layer 2 Claude-only |
sdk-pairing | Agent SDK helpers |
codex-pool | Restart-is-switch, no Layer 2 |
credentials-hygiene | Identity and secret rules |
safe-contribution | Tests, PR window, public-repo hygiene |
Skills tell agents when to call which MCP tool and what never to do. Coding agents still edit the repo with normal tools. Policy skills stop banned patterns from coming back.