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 Arlenjim/claude-code-for-non-coders --skill framing-vague-requestsgit clone --depth 1 https://github.com/Arlenjim/claude-code-for-non-codersWrote 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/arlenjim/claude-code-for-non-coders/framing-vague-requests)<a href="https://agentmods.dev/skills/arlenjim/claude-code-for-non-coders/framing-vague-requests"><img src="https://agentmods.dev/badge/skills/arlenjim/claude-code-for-non-coders/framing-vague-requests/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/arlenjim/claude-code-for-non-coders/framing-vague-requests"><img src="https://agentmods.dev/badge/skills/arlenjim/claude-code-for-non-coders/framing-vague-requests.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 | $0.00051 | $0.01421 |
| Opus 5 | $0.00026 | $0.00711 |
| Sonnet 5 | $0.00010 | $0.00284 |
| Haiku 4.5 | $0.00005 | $0.00142 |
Grade A, and why
framing-vague-requests 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 10d 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 — 64 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Framing a vague request
A badly framed task is lost before the first line: you will quickly produce something clean that doesn't answer the question. Framing is not paperwork — it's the moment where most of the value is decided.
1. Detect that the request is vague
Treat as vague any request where one of these signals appears:
- A goal without an observable criterion: "improve", "clean up", "make it more robust", "modernize". Improved by what measure, visible where?
- A solution given in place of a problem: "add a cache" may be the right answer to a slowness problem — or not. Climb back up to the problem before executing the proposed solution.
- A scope without an edge: you can't say what is explicitly EXCLUDED from the request.
- Terms that admit several readings: "the users", "the payment module", "the old system" — which ones, exactly?
- A gap between the request and what you see in the code: the request describes a state of the system that doesn't match what exists.
If no signal appears — the request is precise, bounded, verifiable — don't inflict framing on it. Execute. Framing a clear request wastes everyone's time.
2. Establish the factual map BEFORE asking questions
Never ask a question the system can answer. Before any question to the requester, go read what exists: what the code does today, what is already wired, what is dead, what the real usages are. Every question whose answer was readable in the code chips away at your credibility and the requester's time — and exposes you to a from-memory answer, often wrong, that you will then execute.
Open with the map, not with a choice. First present the factual state of things: "here's what exists, here's how it's wired, here's what is dead or has drifted". No options at this stage, no recommendation — facts. Two effects: the requester corrects your understanding BEFORE you build on it, and half the questions answer themselves once the facts are laid out.
Flag the gaps between the story and the reality. If the request assumes something the code contradicts ("button X" that doesn't exist, "the old behavior" that never was that), say it first. Building on a false story produces disposable work, even when executed perfectly.
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.
- 10d ago First seen · 64 lines · 51 tokens per session scan A 010c142073c1
framing-vague-requests is a skill published in the GitHub repository Arlenjim/claude-code-for-non-coders (10 stars, last pushed 23d ago), licensed MIT. It adds 51 tokens to every session and 1,421 once invoked, about $0.0003 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.
Other skills, from other repositories
prismstack
A routing guide and builder for Prismstack, a collection of interactive skills for work in a specific field. It helps choose the right skill and load its instructions.
domain-plan
A planning tool for designing a complete set of coding-agent skills for a specific field, including how the skills and their outputs connect.
super-guide
A hands-on guide for learning how a collection of coding-agent skills works. It teaches through the user's own skills and work rather than through abstract examples.
domain-build
A builder that creates a complete collection of coding-agent skills for a specific field from a prepared skill map. It also tracks progress and can continue an interrupted build.
skill-check
A quality-review skill for checking individual skills or an entire skill collection. It uses design questions, review dimensions, and checks for common problems.
source-convert
A translator that turns outside material into reusable coding-agent skill instructions. Sources can include repositories, prompts, videos, articles, books, procedures, codebases, or user ideas.