A code-based way to brainstorm technical changes by putting a rough proposal on a Git branch, where Git records versions of files and changes can be reviewed before approval.

Use it to sketch approaches, explore technical designs, and collect decisions and open questions on an idea branch before promoting the proposal to normal work.

It replaces long proposal text and chat snippets with a reviewable change set. Feedback stays attached to the evolving idea, and nothing becomes a real change until it is approved.

idea — Skill for Claude Code by optimiziramsi/skills

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.

Any agent

npx skills add optimiziramsi/skills --skill idea

Clone the repo

git clone --depth 1 https://github.com/optimiziramsi/skills

Made for: Claude Code.

Or install code-first, the plugin that ships this one along with the rest of its 1 skill, 1 command, 2 hooks.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/optimiziramsi/skills/idea/github.svg)](https://agentmods.dev/skills/optimiziramsi/skills/idea)

Your own site

<a href="https://agentmods.dev/skills/optimiziramsi/skills/idea"><img src="https://agentmods.dev/badge/skills/optimiziramsi/skills/idea/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/optimiziramsi/skills/idea"><img src="https://agentmods.dev/badge/skills/optimiziramsi/skills/idea.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 133 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,235 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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 5.1	$0.00133	$0.01235
Opus 5.5	$0.00053	$0.00494
Sonnet 5.5	$0.00027	$0.00247
Haiku 4.5	$0.00013	$0.00123

Measured today against content hash a4dbb46e3cde, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

idea 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.

code-first/skills/idea/SKILL.md · 82 lines

How it starts

The opening of the file, as written. The whole thing — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.

idea — the proposal is a branch diff

For anything technical, write the idea as code. The user reads it in their review surface as a code review — not chat prose, not inline snippets, not the chat diff view. Rough, first-cut or maybe-wrong is fine: a diff gets a reaction faster than prose. It is still a brainstorm, so nothing lands without the project's normal gate.

Cadence: (1) write the idea as commits on the idea branch; chat holds only the branch name, the key decisions, and the open questions with a recommendation on each. (2) The user reviews the diff and endorses or pushes back — push-back means more idea commits, not a prose rebuttal. (3) On endorse, promote turns it into real commits on a topic branch, which go through the normal gate. No prose-only "here's the plan" turn unless the user asked for one.

Every value below comes from the code-first CLI on PATH (the plugin's bin/), so this skill and the plugin's hooks agree. Every git command here passes git-guard as shipped (no push, no non-FF merge, nothing onto the protected branch) and commit's single-line format.

`/idea <topic>` — start one

code-first base → the branch ideas are cut from. Exit 1 means no safe base (it resolved to a protected branch, or nothing resolved): stop and ask which branch to cut from; suggest setting CODE_FIRST_BASE_BRANCH in the project's settings env.
git status --porcelain — if the tree is dirty, stop and ask: commit, stash, or leave it. Never carry someone's uncommitted work onto an idea branch.
git switch -c "$(code-first prefix)<topic>" <base> — <topic> is short kebab-case.
Write the idea as code: real files where they would live, stubs and TODO(idea): markers are fine, tests if they make the shape clearer. Stage by name.
Commit, one per coherent piece: git commit -m "$(code-first tag): <imperative summary>" — one line, no body (the tag is how idea commits read at a glance; promote drops it).
Reply with: the branch name, a git log --oneline <base>.. count, the 1–3 key decisions, and the open questions — each with your recommendation. No code in chat.

Read the full file on GitHub · 82 lines

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.

today Changed a4dbb46e3cde
yesterday Changed · +1 lines · +23 tokens per session 4ed4677fec07
26d ago First seen · 81 lines · 110 tokens per session scan A d3611eb7b677

Subscribe to this mod's changes

idea is a skill published in the GitHub repository optimiziramsi/skills (2 stars, last pushed today), licensed MIT. It adds 133 tokens to every session and 1,235 once invoked, about $0.0005 per session on Opus 5.5. 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 2026-10-19.

Commits Git and workflow Planning

idea next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
ideaoptimiziramsi/skills	Skill	2	today	MIT	A	133	Claude Code
create-prprisma/orm	Skill	48k	today	Apache-2.0	A	47	Claude Code, Codex
git-commitThinkInAIXYZ/deepchat	Skill	6.4k	yesterday	Apache-2.0	A	15	Claude Code, Codex
engineer-bot-opsmakecindy/cindy	Skill	10k	today	Apache-2.0	A	57	Claude Code, Codex
mcore-split-prNVIDIA/skills	Skill	4.0k	yesterday	Apache-2.0	A	25	Claude Code
contributingHugoRCD/evlog	Skill	1.9k	7d ago	MIT	A	68	Claude Code, Codex
git-commit-writermergisi/awesome-openclaw-agents	Skill	4.1k	10d ago	MIT	A	53	Claude Code, Codex