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 skills add Open-Agent-Tools/General-OAT-Skills --skill scope-itgit clone --depth 1 https://github.com/Open-Agent-Tools/General-OAT-SkillsWrote 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/skills/open-agent-tools/general-oat-skills/scope-it)<a href="https://agentmods.dev/skills/open-agent-tools/general-oat-skills/scope-it"><img src="https://agentmods.dev/badge/skills/open-agent-tools/general-oat-skills/scope-it/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.
<a href="https://agentmods.dev/skills/open-agent-tools/general-oat-skills/scope-it"><img src="https://agentmods.dev/badge/skills/open-agent-tools/general-oat-skills/scope-it.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00145 | $0.01978 |
| Opus 5.5 | $0.00058 | $0.00791 |
| Sonnet 5.5 | $0.00029 | $0.00396 |
| Haiku 4.5 | $0.00015 | $0.00198 |
Grade B, and why
scope-it scanned grade B with 1 finding 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 29d 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.
Asks the agent to reveal its instructionsmediumSystem prompt leakage
Directions to print, repeat or translate the system prompt extract configuration the operator did not intend to expose.
single line of the output prompt. Aim for the fewest questions that fully scope the How it starts
The opening of the file, as written. The whole thing — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Scope It
Turn a vague request into a prompt someone can hand to Claude or Codex and get the right thing back on the first try. The task can be anything — code, writing, research, a decision, a process. You get there by interviewing the user, one question at a time, until four things are pinned down: what, why, constraints, and done.
The analysis tree
Those four are the roots. Under each root hang the decisions that dimension actually depends on — you build that tree from the user's opening message, not from a checklist.
The roots resolve in dependency order, because each one constrains the next:
- Why — the problem, who has it, what breaks if nothing changes. Everything else is unanswerable while this is vague, because "is X in scope?" has no answer without knowing what pain X relieves.
- What — the goal and its boundary. Which outcomes count, which adjacent things are explicitly not being asked for.
- Constraints — the guardrails: whatever must not move, compatibility, budget, deadline, audience, taste, things previously tried that failed.
- Done — the success criteria and the quality bar, stated so that someone else could check them without asking a follow-up question.
The frontier is every unsettled decision whose prerequisites are already answered. Ask from the frontier only. A question whose answer depends on something still open belongs to a later round. Each answer reshapes the tree: it settles a node, opens the ones below it, and sometimes prunes a whole branch as irrelevant.
Running the interview
Find facts yourself; ask only for decisions. Anything that can be looked up, you look up before asking. On this machine that means the filesystem and the shell — what language a project is in, whether a test suite exists, which version is pinned, what already lives in a folder. Off it, that means the web: a library's current API, a service's limits or pricing, a published spec, what a named tool actually does. Reach for either without waiting to be told to. Dispatch subagents for the slower lookups and keep interviewing while they run. Only judgment calls, preferences, tradeoffs, and things that live in the user's head are worth a question.
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.
- 29d ago First seen · 163 lines · 145 tokens per session scan B dbd1a56467df
scope-it is a skill published in the GitHub repository Open-Agent-Tools/General-OAT-Skills (3 stars, last pushed 26d ago), licensed Apache-2.0. It adds 145 tokens to every session and 1,978 once invoked, about $0.0006 per session on Opus 5.5. A static security scan graded it B with 1 finding (asks the agent to reveal its instructions). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-17.
Other skills, from other repositories
remotion-video
Make a short video about a product or brand (promo, launch, explainer, demo or social clip) entirely in code with Remotion, from idea to a rendered MP4 with sound. The owner approves the idea and a storyboard before anything is built.
mvp-plan
Plan the MVP build from a PRD and an MVP scope, as a full development plan or an ultra-lean build spec, with the AI Product Development Toolkit's planning prompts.
user-research
Plan a user research questionnaire, or turn collected responses into evidence-linked input for a PRD, with the AI Product Development Toolkit's research prompts.
prd
Interview the owner about a product idea, then draft a Product Requirements Document (PRD) with the AI Product Development Toolkit's PRD prompt.
nextjs-supabase-vercel
Set up the current empty folder with the AI App Starter, so the agent builds a Next.js, Supabase and Vercel app with the owner step by step, from idea to launch.
handoff
Hand work to another tool, account or machine and pick it up there, in two words. Use when the user says "hand off", "hand off all", "out of tokens", "switching accounts", "pick up" or "continue".