A setup skill that adds an ICM project structure to a repository, including guidance files, folders for specifications and plans, and a change log. It leaves existing files unchanged where possible.

Use it to create the ICM tree, update missing ignore rules, and connect CLAUDE.md to AGENTS.md after checking that uv is available.

It gives a repository a consistent place for project context, requirements, plans, and implementation records.

init — Skill by andyrids

Install

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.

agentmods

npx agentmods add skills/andyrids/icm-spec/init

Any agent

npx skills add andyrids/icm-spec --skill init

Clone the repo

git clone --depth 1 https://github.com/andyrids/icm-spec

Made for: Claude Code, Codex.

Or install icm, the plugin that ships this one along with the rest of its 8 skills, 1 agent, 5 hooks.

Per session 37 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 652 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00037	$0.00652
Opus 5	$0.00018	$0.00326
Sonnet 5	$0.00007	$0.00130
Haiku 4.5	$0.00004	$0.00065

Measured 2d ago against content hash 6ed3dc408245, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

init 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.

skills/init/SKILL.md · 52 lines

What it actually says

Scaffold ICM into this repository

First, run uv --version through Bash. If it fails, stop and scaffold nothing: every gate this plugin ships runs through uv, and a gate that cannot start is treated as a non-blocking error, so the tree would carry enforcement that silently fails open. Tell the user to install uv (https://docs.astral.sh/uv/getting-started/installation/) and re-run /icm:init.

Then copy the tree under ${CLAUDE_SKILL_DIR}/templates/ into the repository root. The operation is idempotent: a destination file that already exists is left untouched, and re-running reports exists per file rather than rewriting anything.

Mapping, template path to destination:

Template	Destination
`AGENTS.md`	`AGENTS.md`
`CONTEXT.md`	`CONTEXT.md`
`CHANGELOG.md`	`CHANGELOG.md`
`gitignore`	`.gitignore`
`specs/README.md`	`specs/README.md`
`plans/README.md`	`plans/README.md`
`ICM/**` (all files)	`ICM/**`

Then:

If .gitignore already existed, append any template lines it is missing (the stage output/ ignore is load-bearing) instead of overwriting.
Point CLAUDE.md at AGENTS.md: create a symlink where the platform allows it, otherwise write a CLAUDE.md containing only @AGENTS.md.
Copy every output/.gitkeep and shared/.gitkeep so the empty directories survive a clone.
Report one line per file - written or exists - and finish by naming the two entry points: /icm:specify <feature request> for work that changes what specs/** declares, and /icm:express <change request> for work that conforms to a spec already committed. The root CONTEXT.md is the authority on choosing between them.

Do not tailor the copied files to the project during init. They are the factory configuration; project-specific reference material (ICM/_config/reference-toolchain-*.md) is added by the user as tools come into play.

The one exception is the [Project name] identity block at the top of AGENTS.md, which is a placeholder rather than configuration. Leave it as written and tell the user it is theirs to fill in - an unfilled identity block is visible, whereas a guessed one is not.

Files

What ships with it

25 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

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 · 52 lines · 37 tokens per session scan A 6ed3dc408245

Subscribe to this mod's changes

init is a skill published in the GitHub repository andyrids/icm-spec (2 stars, last pushed 14d ago), licensed Apache-2.0. It adds 37 tokens to every session and 652 once invoked, about $0.0002 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.

agent-architecture claude-code claude-plugin icm interpretable-context-methodology spec-driven-development

tldr

Generate a review-oriented TLDR of GitHub Spec Kit artifacts (spec.md and/or plan.md) as a self-contained HTML dashboard plus a Markdown version, to make reviewing those files in a pull request faster. Use this skill whenever the user wants to review, summarize, "TLDR", skim, or get an overview of a Spec Kit spec.md…

qurore/speckit-tldr · 141 tokens

tldr-delete

Remove generated speckit-tldr output — the .tldr.html and .tldr.md files and the specs/ /tldr/ directories that hold them — so these local review artifacts are not committed into a pull request. Use this skill whenever the user wants to delete, remove, clean, clear, or "clean up" the TLDRs that the tldr skill…

qurore/speckit-tldr · 123 tokens

create-mcp-server

Build, iterate, and evaluate Model Context Protocol (MCP) servers that expose external services as tools an LLM can call. Covers schema/tool design, error handling, pagination, MCP Inspector testing, and an eval set. Use when asked to "build an MCP server", "create an MCP tool", "wrap this API as MCP", "expose X to…

open-gsd/gsd-pi · 90 tokens

model-compatibility

Model family compatibility matrix covering loaders, resolutions, samplers, CFG, VAE, ControlNet, and LoRA compatibility for SD 1.5, SDXL, Flux, SD3, and video models.

artokun/comfyui-mcp · 47 tokens

comfyui-node-registry

Authoring & publishing ComfyUI custom nodes to the Comfy Registry, covering node structure, pyproject.toml spec, comfy-cli publishing, and CI.

artokun/comfyui-mcp · 38 tokens

civitai

Discover Civitai models with the BUILT-IN downloadmodel action:"searchcivitai" and install/generate them locally. Find a checkpoint/LoRA/embedding on Civitai, download it into ComfyUI, and use its trigger words. Optionally pair the official Civitai MCP for community features (images browsing, posting, collections).

artokun/comfyui-mcp · 76 tokens