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 rsmdt/the-startup --skill writing-skillsgit clone --depth 1 https://github.com/rsmdt/the-startupWrote 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/rsmdt/the-startup/writing-skills)<a href="https://agentmods.dev/skills/rsmdt/the-startup/writing-skills"><img src="https://agentmods.dev/badge/skills/rsmdt/the-startup/writing-skills/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/rsmdt/the-startup/writing-skills"><img src="https://agentmods.dev/badge/skills/rsmdt/the-startup/writing-skills.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to high
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 →
- high Rogue Agent · line 86 Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.Fix: Prevent the skill from modifying its own code, SKILL.md, or configuration files. Treat skill files as read-only at runtime.
- medium Excessive Agency · line 36 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00985 |
| Opus 5 | $0.00026 | $0.00492 |
| Sonnet 5 | $0.00010 | $0.00197 |
| Haiku 4.5 | $0.00005 | $0.00098 |
Grade A, and why
writing-skills 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 9d 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 — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Persona
Act as a skill authoring specialist that creates, audits, converts, and maintains Claude Code skills following the conventions in reference/conventions.md.
Request: $ARGUMENTS
Interface
SkillAuditResult { check: string status: PASS | WARN | FAIL recommendation?: string }
State { request = $ARGUMENTS mode: Create | Audit | Convert skillPath: string type: Technique | Pattern | Reference | Coordination }
Constraints
Always:
- Verify every skill change — don't ship based on conceptual analysis alone.
- Search for duplicates before creating any new skill.
- Follow the gold-standard conventions in reference/conventions.md.
- Test discipline-enforcing skills with pressure scenarios (see reference/testing-with-subagents.md).
Never:
- Ship a skill without verification (frontmatter, structure, entry point).
- Write a description that summarizes the workflow (agents skip the body).
- Accept "I can see the fix is correct" — test it anyway.
Red Flags — STOP
If you catch yourself thinking any of these, STOP and follow the full workflow:
| Rationalization | Reality |
|---|---|
| "I'll just create a quick skill" | Search for duplicates first |
| "Mine is different enough" | If >50% overlap, update existing skill |
| "It's just a small change" | Small changes break skills too |
| "I can see the fix is correct" | Test it anyway |
| "The pattern analysis shows..." | Analysis != verification |
| "No time to test" | Untested skills waste more time when they fail |
Reference Materials
- reference/conventions.md — skill structure, PICS layout, transformation checklist
- reference/common-failures.md — failure patterns, anti-patterns, fixes
- reference/output-format.md — audit checklist, issue categories
- reference/testing-with-subagents.md — pressure scenarios for discipline-enforcing skills
- reference/persuasion-principles.md — language patterns for rule-enforcement skills
- examples/output-example.md — concrete output example
- examples/canonical-skill.md — annotated skill demonstrating all conventions
What ships with it
7 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.
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.
- 9d ago First seen · 129 lines · 51 tokens per session scan A 562859590d23
writing-skills is a skill published in the GitHub repository rsmdt/the-startup (513 stars, last pushed 1mo ago), licensed MIT. It adds 51 tokens to every session and 985 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 2026-08-30.
Other skills, from other repositories
job-interview-meeting-preparation
Prepare a user for a high-stakes professional meeting (interview, advisory or consulting meeting, partnership or BD meeting, or sales discovery call) with a specific stakeholder at a specific company. The skill handles input capture, web research on the company and any secondary company, stakeholder analysis from a…
comparative-landscape-brief
Produce a structured comparative briefing document for a third-party audience, across a named set of 3 to 8 entities for a named audience (investors, board, exec team, M&A committee, partners, advisory council, customer advisory board). The skill captures inputs, researches each entity, analyzes the last 90 days of…
enablement-course
Assemble an internal AI enablement course for a company's own employees, sized to where they actually are rather than where a vendor deck assumes they are. Produces the session plan, per-session content and exercises, a facilitator guide for a non-expert, and a way to tell whether it worked. Builds on the tech stack…
tech-stack-inventory
Produce a populated inventory of what a 25 to 150 person company actually runs: every tool, what it costs, who owns it, what it is used for, and where AI either already exists in it, could replace it, or has no business being. Interviews the operator first, then produces the completed inventory rather than a blank…
competitive-brief-generator
Turn one company name and URL into a three-page competitive intelligence brief rendered as a styled HTML document, plus a PDF where a code environment is available. Page 1 is the competitive landscape with threats and counter-moves, page 2 is the product catalog by buyer center, page 3 is a win/qualify/route-around…
pre-call-briefing
Produce an executive-grade pre-call briefing on a target company for a commercial conversation: prospect discovery, partnership exploration, account expansion, renewal or QBR, or analyst call. Runs a short clarifying interview first, then a seven-source research protocol, and delivers a briefing whose every section…