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.
git clone --depth 1 https://github.com/ssheleg/make-skillWrote 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/rules/ssheleg/make-skill/skill-search)<a href="https://agentmods.dev/rules/ssheleg/make-skill/skill-search"><img src="https://agentmods.dev/badge/rules/ssheleg/make-skill/skill-search/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/rules/ssheleg/make-skill/skill-search"><img src="https://agentmods.dev/badge/rules/ssheleg/make-skill/skill-search.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 | $0.00117 | $0.00499 |
| Opus 5.5 | $0.00047 | $0.00200 |
| Sonnet 5.5 | $0.00023 | $0.00100 |
| Haiku 4.5 | $0.00012 | $0.00050 |
Grade B, and why
skill-search 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 5d 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.
Enumerates other installed skillsmediumAgent snooping
Other skills' SKILL.md files reveal prompts, capabilities and secrets that should be invisible to peers.
`find -L ~/.agents/skills ~/.claude/skills ~/.cursor/skills -name SKILL.md -exec grep -ilE "^description:.*(<concept>|<concept>)" {} + 2>/dev/null` What it actually says
skill-search — the first step of every task (Cursor)
Self-contained rule: Cursor copies this file into foreign projects, so it uses absolute URLs only. Source: https://github.com/ssheleg/make-skill.
Most installed skills are hidden from the host's skill listing on purpose, so the listing is not the catalogue. Search the catalogue.
- Restate the request in one line — the meaning, not the literal phrasing.
- Decide whether a skill could help. None for a question, an explanation or a one-line edit: say so and proceed.
- Turn the meaning into 3–8 concepts in English AND Russian, plus synonyms and
the artifact involved (
pdf, документ, extract text, извлечь текст). - Search:
npx --yes sshlg-skills toolkit --find "<concept>, <concept>, …"— it prints id, score, matched concepts, visibility and the SKILL.md path. Hidden matches count exactly as listed ones. - No node/network, or the launcher does not know
--find(older than 1.55.0):find -L ~/.agents/skills ~/.claude/skills ~/.cursor/skills -name SKILL.md -exec grep -ilE "^description:.*(<concept>|<concept>)" {} + 2>/dev/null— once; nothing found means no skill, never a loop. - Open the SKILL.md of what fits, read it, follow it, and say in one line which skill you took (or that none applied).
Building, auditing or publishing a skill is the make-skill rule, not this one.
A third-party skill nobody reviewed gets an install review before its scripts run.
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.
- 5d ago First seen · 30 lines · 117 tokens per session scan B 794381143528
skill-search is a cursor rule published in the GitHub repository ssheleg/make-skill (4 stars, last pushed 3d ago), licensed MIT. It adds 117 tokens to every session and 499 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it B with 1 finding (enumerates other installed skills). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-10.
Other cursor rules, from other repositories
archcore-context
Archcore knowledge base context — document types, MCP tools, and conventions for working with .archcore/ documents.
archcore-files
Enforce MCP-only operations when working with .archcore/ files.
digital-brain
A Claude Code plugin and Obsidian vault that builds persistent, compounding knowledge bases using Andrej Karpathy's LLM Wiki pattern. This repo works with Cursor's AI through the cross-platform Agent Skills format.
ponytail
Ponytail, lazy senior dev mode. Always pick the simplest solution that works.
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.
dev-standard
Apache Superset development standards and guidelines for Cursor IDE.