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 naveedharri/benai-skills --skill prompt-mastergit clone --depth 1 https://github.com/naveedharri/benai-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/naveedharri/benai-skills/prompt-master)<a href="https://agentmods.dev/skills/naveedharri/benai-skills/prompt-master"><img src="https://agentmods.dev/badge/skills/naveedharri/benai-skills/prompt-master/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/naveedharri/benai-skills/prompt-master"><img src="https://agentmods.dev/badge/skills/naveedharri/benai-skills/prompt-master.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Memory Poisoning · line 47 Skill injects content designed to persist in agent memory or context across interactions. Persistent injection can alter agent behavior long after the initial interaction.Fix: Do not allow untrusted input to persist in agent memory or context. Validate all content before storing and implement memory isolation between sessions.
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 | $0.00225 | $0.01289 |
| Opus 5 | $0.00112 | $0.00645 |
| Sonnet 5 | $0.00045 | $0.00258 |
| Haiku 4.5 | $0.00022 | $0.00129 |
Grade A, and why
prompt-master 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 6d 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 — 63 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Prompt Master
Turns what the user wants into a prompt, brief or instruction file written the way Anthropic says to prompt the Claude 5 generation. Always begins by asking. The rulebook with sources is references/rulebook.md; read it once per session, then work from the steps.
Steps
Track progress:
Task Progress:
- [ ] 1. Interview
- [ ] 2. Strip what is retired
- [ ] 3. Rebuild as a Full Job Brief (or audit the file)
- [ ] 4. Rules to reasons
- [ ] 5. Deliver with receipts
1. Interview
Read what the user gave you first (the pasted prompt, the file, the folder). Then interview one question at a time following references/interview.md: never ask what you could read, prioritise questions whose answer changes the shape of the prompt, stop at 7. Skip to a single confirming question when the ask is already small and clear. By the end you know the input type (rough idea, existing prompt, or whole instruction file), the surface (Claude Cowork, Claude app, Claude Code, API), the run length (quick answer or long run), who the output is for, and what done looks like.
2. Strip what is retired
For an existing prompt or file, check every line against references/retired-instructions.md; remove or replace what matches and record each change for step 5. For a rough idea, skip this step.
3. Rebuild as a Full Job Brief, or audit the file
- Rough idea or existing prompt: write it in the four-part shape from
references/job-brief.md(job, why, guardrails, done-means). For a long run, append the audit line verbatim from that file. Drop any part the task does not need; a three-line prompt is right for a three-line task. - Whole instruction file (CLAUDE.md, skill, project instructions): run the bucket audit in
references/file-audit.md, produce the three-column review table, and wait for approval before editing anything.
4. Rules to reasons
Re-read the draft or the proposed rewrites. Any hard rule without a reason gets the reason or gets cut, using references/rules-to-reasons.md. Keep hard rules only where being wrong is expensive. Say what to do instead of what not to do.
What ships with it
9 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.
- EVAL.md 1.2 KB
- references/examples/client-report-rewrite.md 2.8 KB
- references/file-audit.md 2.1 KB
- references/interview.md 2.4 KB
- references/job-brief.md 3.2 KB
- references/retired-instructions.md 3.0 KB
- references/rulebook.md 3.7 KB
- references/rules-to-reasons.md 2.0 KB
- references/voice-and-format.md 2.4 KB
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.
- 6d ago First seen · 63 lines · 225 tokens per session scan A 2513c92d6d9c
prompt-master is a skill published in the GitHub repository naveedharri/benai-skills (61 stars, last pushed 8d ago), licensed MIT. It adds 225 tokens to every session and 1,289 once invoked, about $0.0011 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 2026-09-05.
Other skills, from other repositories
ai-prompting-basics
A practical guide to writing clearer requests for Claude, an AI assistant, by adding context, examples, roles, and output rules.
llm-router-ops
Operate a self-hosted OpenAI-compatible LLM router/proxy (e.g. LiteLLM) in front of one or more backends — the minimal client-wiring block for every client type, the context-window advertisement gotcha, the env-vs-persisted-config gotcha, and why an unauthenticated health probe should 401, not 200. Use when wiring a…
guidance
Constrain LLM output with grammars; guarantee valid JSON.
outlines
Outlines: structured JSON/regex/Pydantic LLM generation.
instructor
Structured LLM outputs validated with Pydantic.
dspy
DSPy: declarative LM programs, auto-optimize prompts, RAG.