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/controlplane-com/ai-plugin/claude-mdgit clone --depth 1 https://github.com/controlplane-com/ai-pluginWhat 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 | $0.01995 | $0.01995 |
| Opus 5 | $0.00997 | $0.00997 |
| Sonnet 5 | $0.00399 | $0.00399 |
| Haiku 4.5 | $0.00199 | $0.00199 |
Grade A, and why
ai-plugin CLAUDE.md scanned grade A with 0 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 101 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Contributor context — Claude Code
End-user install and capability docs live in README.md. Development principles, validation, and release process live in CONTRIBUTING.md. This file is the working context for Claude Code sessions on this repo.
Repo layout
Everything lives under plugins/cpln/. All skills, rules, commands, agents, hooks, and per-client manifests are authored there, and every client (Claude Code, Codex, Cursor, Antigravity CLI) resolves that single directory as its plugin root.
| Path | Purpose |
|---|---|
plugins/cpln/ |
The plugin itself. Claude, Codex, Cursor, and Antigravity CLI all resolve this as their plugin root. |
plugins/cpln/skills/<name>/SKILL.md |
One domain skill per folder. Companion files (*.md) load on demand. |
plugins/cpln/agents/<name>.md |
One self-contained guided workflow per file. |
plugins/cpln/commands/<name>.md |
Slash command for Claude / Codex / Cursor. |
plugins/cpln/rules/*.md |
Guardrails and manifest references. Files with alwaysApply: true are injected by the SessionStart hook in plugins/cpln/hooks/hooks.json (Claude / Codex) and read by Cursor as native rules. Antigravity has no SessionStart hook and does not load a plugin rules/ dir (its rules live in AGENTS.md), so it gets the same guardrails via the MCP server (get_cpln_rules). |
plugins/cpln/.claude-plugin/plugin.json |
Claude plugin manifest. |
plugins/cpln/.codex-plugin/plugin.json + mcp.json |
Codex manifest and MCP config. |
plugins/cpln/.cursor-plugin/plugin.json + mcp.json |
Cursor manifest and MCP config. |
plugins/cpln/plugin.json + mcp_config.json |
Native Antigravity CLI (agy) manifest and MCP config (remote server uses serverUrl, not httpUrl). Antigravity has no SessionStart hook event (its hooks are PreToolUse/PostToolUse/PreInvocation/PostInvocation/Stop), so guardrails reach it through the MCP server (get_cpln_rules + server-side skill gate), not a hook. agy plugin validate ./plugins/cpln gates it; install with agy plugin install https://github.com/controlplane-com/ai-plugin/plugins/cpln. |
plugins/cpln/.claude-mcp.json |
Claude MCP config. |
.claude-plugin/marketplace.json |
Claude marketplace entry. Source: "./plugins/cpln". |
.agents/plugins/marketplace.json |
Codex marketplace entry. |
.cursor-plugin/marketplace.json |
Cursor marketplace entry. |
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.
- 2d ago First seen · 101 lines · 1,995 tokens per session scan A 286881d12f62
ai-plugin CLAUDE.md is an instructions file published in the GitHub repository controlplane-com/ai-plugin (10 stars, last pushed 8d ago), licensed MIT. It adds 1,995 tokens to every session, about $0.0100 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
rustfs AGENTS.md
Instructions for rustfs/rustfs, covering rustfs agent instructions, precedence, operating model, worktree and disk hygiene and change style.
rustfs CLAUDE.md
Instructions for rustfs/rustfs, covering rustfs — claude.md, commands, where to look (do not duplicate here) and domain conventions worth knowing up front.
camel AGENTS.md
AGENTS.md instructions for apache/camel, covering apache camel - ai agent guidelines, project info, ai agent rules of engagement, attribution and pr volume.
apisix AGENTS.md
Instructions for apache/apisix, covering apache apisix — agent instructions and security model.
graphjin AGENTS.md
AGENTS.md instructions for dosco/graphjin, covering graphjin agent guide, architectural overview, directory structure & responsibilities, build commands and coding guidelines.
higress AGENTS.md
AGENTS.md instructions for higress-group/higress, covering agents.md, mandatory issue-spec gate for agent-assisted changes, repository layout, plugins and plugins/wasm-go/ (primary wasm plugin framework, go).