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/comonicon/comonicon.jl/agents-mdgit clone --depth 1 https://github.com/comonicon/Comonicon.jlWrote 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/comonicon/comonicon.jl/agents-md)<a href="https://agentmods.dev/instructions/comonicon/comonicon.jl/agents-md"><img src="https://agentmods.dev/badge/instructions/comonicon/comonicon.jl/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 5.1 | $0.00840 | $0.00840 |
| Opus 5 | $0.00420 | $0.00420 |
| Sonnet 5 | $0.00168 | $0.00168 |
| Haiku 4.5 | $0.00084 | $0.00084 |
Grade A, and why
Comonicon.jl 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 today.
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 — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Project
Comonicon is a Julia CLI generator that transforms function definitions and docstrings into full-featured command-line interfaces with zero duplication, zero overhead, and zero dependencies. The two primary user-facing macros are @cast (mark a function/module as a CLI command) and @main (declare the CLI entry point).
Build & Test Commands
julia --project -e 'using Pkg; Pkg.instantiate()' # Install dependencies
julia --project test/runtests.jl # Run all tests
julia --project -e 'using Pkg; Pkg.test()' # Run tests via Pkg
To run a single testset:
julia --project -e 'using Test; include("test/runtests.jl")' # all
# Run a specific file directly (most test files are self-contained):
julia --project test/options.jl
julia --project test/frontend/cast.jl
Pre-commit Checklist
julia --project -e 'using JuliaFormatter; format("src")' # Format (margin=102)
julia --project test/runtests.jl # Tests
JuliaFormatter config: always_for_in=true, margin=102.
Architecture
The pipeline is: Julia source → Frontend (parse) → AST → Codegen → executable CLI.
Modules
| Module | Path | Role |
|---|---|---|
| Frontend | src/frontend/ |
@cast/@main macros; parse function signatures and markdown docstrings into CLI metadata |
| AST | src/ast/ |
Immutable command-tree types: Entry → NodeCommand/LeafCommand → Argument/Option/Flag |
| Codegen | src/codegen/ |
emit() converts Entry to Julia Expr; also generates bash/zsh completions |
| Builder | src/builder/ |
Compilation, installation, system image generation via PackageCompiler |
Key Types (src/ast/types.jl)
Entry
└─ root: Union{NodeCommand, LeafCommand}
├─ NodeCommand — routing node; holds OrderedDict of subcommands
└─ LeafCommand — executable leaf; holds Vector[Argument], OrderedDict[Option/Flag]
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.
- today First seen · 76 lines · 840 tokens per session scan A 1f56f450ab40
Comonicon.jl AGENTS.md is an instructions file published in the GitHub repository comonicon/Comonicon.jl (298 stars, last pushed 2mo ago), licensed MIT. It adds 840 tokens to every session, about $0.0042 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 2026-09-04.
Other instructions, from other repositories
cli AGENTS.md
Instructions for axiomhq/cli, covering agents.md, build & development commands, architecture, entry point and factory pattern.
smbcloud-cli copilot-instructions.md
Copilot instructions for smbcloudXYZ/smbcloud-cli, covering github copilot instructions — smbcloud-cli, project overview, workspace structure, cli source layout (crates/cli/src/) and dependency graph.
vulcano AGENTS.md
AGENTS.md instructions for dgarana/vulcano, covering agents.md, project purpose, project vision, working rules for ai agents and quality guardrails.
smbcloud-cli AGENTS.md
AGENTS.md instructions for smbcloudXYZ/smbcloud-cli, a project described as: CLI and MCP server for smbCloud: email service on smbCloud Mail, authentication apps on smbCloud Auth, and app deploys. One Rust binary.
smbcloud-cli CLAUDE.md
Claude Code instructions for smbcloudXYZ/smbcloud-cli, a project described as: CLI and MCP server for smbCloud: email service on smbCloud Mail, authentication apps on smbCloud Auth, and app deploys. One Rust binary.
cli CLAUDE.md
Instructions for axiomhq/cli, a project described as: The power of Axiom on the command line.