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/greenpolo/cc-multi-cli-plugin/agents-mdgit clone --depth 1 https://github.com/greenpolo/cc-multi-cli-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 | $0.01848 | $0.01848 |
| Opus 5 | $0.00924 | $0.00924 |
| Sonnet 5 | $0.00370 | $0.00370 |
| Haiku 4.5 | $0.00185 | $0.00185 |
Grade A, and why
cc-multi-cli-plugin AGENTS.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 yesterday.
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 — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Orientation for AI agents (Claude, Codex, Cursor) working in this repo. Read this first.
What this is
cc-multi-cli-plugin is a Claude Code plugin that offloads heavy coding work to external CLIs — Codex, Cursor (headless agent -p), Antigravity, and OpenCode (headless opencode run --format json) — so the orchestrating Claude session spends as few tokens as possible. The plugin's entire value is token reduction by delegation. Keep that goal central to every change.
The golden rule
Claude does as little thinking as possible; the external CLI does the work.
- The per-command subagents (
plugins/multi/agents/codex-*.md) are thin forwarders: they build exactly one companion command, run it, and return stdout unchanged. They must not read files, reason about the task, or "help." - Forwarder models are tuned by role. Forwarders that frame/route the prompt —
codex-execute,codex-rescue, and the cursor/antigravity ones — run on Sonnet, because shaping the task well (model/effort choice, prompt framing) materially improves what the external CLI then produces (this mirrors the officialcodex-plugin-ccrescue subagent). Forwarders that do no framing and only bridge to the companion get no subagent at all: their slash command runs the companion directly from the main loop viaBash(node:*)(/codex:review,/codex:adversarial-review), which is strictly cheaper than a forwarder — this mirrorscodex-plugin-cc'scommands/review.md. Keep the remaining forwarders thin: never add file reading or task reasoning. - Do not spawn fleets of Claude subagents to do or validate work in this repo — that defeats the plugin's purpose. Validate with
npm test(offline) or by delegating to a CLI.
The request chain
/codex:<cmd> → multi:codex-<role> (framing forwarder, Sonnet) — or, for review, no subagent at all
→ multi-cli-companion.mjs <subcommand> --cli <name>
→ lib/adapters/<name>.mjs → (codex app-server broker | cursor headless `agent -p` | antigravity headless `agy -p` | opencode headless `opencode run --format json`) → external CLI
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.
- yesterday First seen · 59 lines · 1,848 tokens per session scan A 8ab7d2692131
cc-multi-cli-plugin AGENTS.md is an instructions file published in the GitHub repository greenpolo/cc-multi-cli-plugin (89 stars, last pushed 16d ago), licensed Apache-2.0. It adds 1,848 tokens to every session, about $0.0092 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 2026-08-30.
Other instructions, from other repositories
intelligent-terminal agent-skills.instructions.md
Guidelines for creating high-quality Agent Skills for GitHub Copilot.
intelligent-terminal localization.instructions.md
Instructions for microsoft/intelligent-terminal, covering localization instructions for .resw resource files, components with localized resources, locale categories, authoritative locale list (from terminalapp) and adding or updating localized strings.
intelligent-terminal rust-localization.instructions.md
Instructions for microsoft/intelligent-terminal, covering localization instructions for rust wta (yaml locale files), file structure, file format, the en-us.yml source file and non-translatable token rules.
intelligent-terminal copilot-instructions.md
Instructions for microsoft/intelligent-terminal, covering copilot instructions for windows terminal, build, running, test and code formatting.
intelligent-terminal instructions.instructions.md
Guidelines for creating high-quality custom instruction files for GitHub Copilot.
intelligent-terminal AGENTS.md
Instructions for microsoft/intelligent-terminal, covering intelligent terminal, architecture, supported agents and settings, user-facing behavior and key files.