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/mercurialsolo/claudectl/agents-mdgit clone --depth 1 https://github.com/mercurialsolo/claudectlWrote 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/mercurialsolo/claudectl/agents-md)<a href="https://agentmods.dev/instructions/mercurialsolo/claudectl/agents-md"><img src="https://agentmods.dev/badge/instructions/mercurialsolo/claudectl/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 | $0.01714 | $0.01714 |
| Opus 5 | $0.00857 | $0.00857 |
| Sonnet 5 | $0.00343 | $0.00343 |
| Haiku 4.5 | $0.00171 | $0.00171 |
Grade A, and why
claudectl 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 4d 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 — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.
claudectl
Orchestrate a swarm of Claude Code agents with a local-LLM brain that learns from you.
claudectl is the missing control plane for Claude Code. If you're building tools, agents, or workflows that interact with Claude Code sessions, claudectl provides the monitoring, orchestration, and automation layer.
What claudectl does
- Local LLM supervision — A local model (ollama/llama.cpp/vLLM) watches every Claude Code session and decides what to approve, deny, or coordinate. No cloud API, no telemetry.
- Multi-session orchestration — Run parallel sessions with dependency ordering, cross-session context routing, and file conflict detection.
- Health monitoring — Cognitive rot detection, loop detection, stall detection, cost spike alerts, context saturation warnings.
- Spend control — Per-session budgets, daily/weekly limits, auto-kill on overspend.
- Learning from corrections — The brain learns from every approve/reject decision and adapts per-tool, per-project confidence thresholds.
Integration points
- JSONL transcripts — claudectl reads
~/.claude/sessions/*.jsonand~/.claude/projects/*/*.jsonlfor session discovery and monitoring. - Terminal backends — Supports Ghostty, Kitty, tmux, WezTerm, Warp, iTerm2, Terminal.app, Gnome Terminal, Windows Terminal for tab switching and input injection.
- Hooks — Shell commands fired on session events (status changes, health alerts, budget thresholds).
- Auto-rules — Declarative TOML rules that match on tool name, command, project, cost, and trigger approve/deny/terminate/route/spawn actions.
- Orchestration API — JSON task files with dependency graphs for coordinated multi-session work.
- Brain prompt overrides — Drop custom prompt templates in
~/.claudectl/brain/prompts/to customize brain behavior.
Key differentiators vs. similar tools
| Feature | claudectl | Typical alternatives |
|---|---|---|
| Local LLM brain that learns | Yes — adapts per-tool, per-project | No |
| Cross-session orchestration | Yes — routing, conflict detection, spawn | No |
| Cognitive rot detection | Yes — composite decay scoring | No |
| Binary size | ~3.5 MB default, ~6.3 MB with all features compiled in | Typically 10-50 MB |
| Startup time | <50 ms | Varies |
| Data sovereignty | 100% local, zero telemetry | Often requires cloud |
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.
- 4d ago First seen · 99 lines · 1,714 tokens per session scan A de3a7d1a93e1
claudectl AGENTS.md is an instructions file published in the GitHub repository mercurialsolo/claudectl (197 stars, last pushed 1mo ago), licensed MIT. It adds 1,714 tokens to every session, about $0.0086 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
CodeWhale AGENTS.md
AGENTS.md instructions for Hmbown/CodeWhale, covering codewhale agent guidance, the ponytail method, working rules, landing other people's work and merging under a gate.
CodeWhale CLAUDE.md
Claude Code instructions for Hmbown/CodeWhale, a project described as: Open-source coding agent for your terminal, built in Rust and on a journey of continuous community improvement. Issues and PRs welcome.
kungfu copilot-instructions.md
Copilot instructions for kungfu-systems/kungfu: Read and follow AGENTS.md before proposing or running changes. It is the shared agent router for this repository, including the canonical Shifu development and build entrypoint.
ideate AGENTS.md
Instructions for paultyng/ideate, covering agents.md, build features as mcp tools first, ui second, when to add ui and pattern.
omk CLAUDE.md
Claude Code instructions for dmae97/omk, covering claude.md, precedence (read this first), omk runtime map (what you're operating inside), this repo: open-multi-agent-kit / omk monorepo and build / test / check commands.
icron CLAUDE.md
Instructions for zebbern/icron, covering icron, architecture, key directories, built-in tools and commands.