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 oleksbard/claude-find-skill --skill find-skillgit clone --depth 1 https://github.com/oleksbard/claude-find-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/skills/oleksbard/claude-find-skill/find-skill)<a href="https://agentmods.dev/skills/oleksbard/claude-find-skill/find-skill"><img src="https://agentmods.dev/badge/skills/oleksbard/claude-find-skill/find-skill/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/oleksbard/claude-find-skill/find-skill"><img src="https://agentmods.dev/badge/skills/oleksbard/claude-find-skill/find-skill.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 | $0.01836 |
| Opus 5 | $0.00059 | $0.00918 |
| Sonnet 5 | $0.00023 | $0.00367 |
| Haiku 4.5 | $0.00012 | $0.00184 |
Grade C, and why
find-skill scanned grade C with 3 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
find ~/.claude/plugins -maxdepth 8 -name SKILL.md 2>/dev/null # plugin skills Enumerates other installed skillsmediumAgent snooping
Other skills' SKILL.md files reveal prompts, capabilities and secrets that should be invisible to peers.
ls -d ~/.claude/skills/*/ 2>/dev/null # personal skills Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -sG "https://api.github.com/search/repositories" \ How it starts
The opening of the file, as written. The whole thing — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Find a Skill
Discover Claude Code skills that are not yet installed and recommend the best matches for a stated need. This is discover-and-recommend only — never install, download, clone, or modify the user's system. End by showing a ranked shortlist with links; the user installs manually.
Procedure
1. Turn the request into keywords
Extract 2–4 search keywords from the user's need (the topic, the verb, key nouns). If the request is too vague to search (e.g. "find me a useful skill"), ask one clarifying question first.
2. List what's already installed (for dedup)
The whole point is to surface skills the user does not have. Build a set of installed skill names/topics:
ls -d ~/.claude/skills/*/ 2>/dev/null # personal skills
find ~/.claude/plugins -maxdepth 8 -name SKILL.md 2>/dev/null # plugin skills
Also account for the available-skills list already present in your context. Any candidate that matches something installed must be excluded — or mentioned as "you already have this" — never recommended as new.
3. Search the sources
Run all three. They are independent; if one fails, continue with the others.
A. Official marketplace — local cache, offline, fast. Always do this first.
jq -r --arg kw "<keyword>" '
.catalog.plugins | to_entries[]
| .key as $id | .value as $p
| ($p.marketplace_entry.name // $id) as $name
| ($p.marketplace_entry.description // "") as $desc
| ([$p.components.skills[]?.name] | join(", ")) as $skills
| select((($name+" "+$desc+" "+$skills)|ascii_downcase) | test($kw|ascii_downcase))
| "\($p.unique_installs // 0)\t\($name)\t\($id)\t\($p.marketplace_entry.homepage // "")\t\($desc)"
' ~/.claude/plugins/plugin-catalog-cache.json | sort -rn | head -10
Columns: unique_installs (popularity) · name · install id (name@marketplace) · homepage · description. Install hint for these: /plugin install <id>.
B. GitHub — degradation chain. Use the first rung that works.
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 · 105 lines · 117 tokens per session scan C bd4c517ee0b5
find-skill is a skill published in the GitHub repository oleksbard/claude-find-skill (10 stars, last pushed 2mo ago), licensed MIT. It adds 117 tokens to every session and 1,836 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it C with 3 findings (reads agent configuration directories, enumerates other installed skills, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
manus-prompt-engineering
Design and audit Manus prompts for engineering, release, production, and security work. Use when writing a Manus prompt or reviewing a Manus execution or recording against authorization and evidence.
ui-craft
You are a principal product designer and senior frontend engineer. Your job is to produce interfaces that feel handcrafted by a specialist — not assembled from AI defaults.
skill-builder
Automatically detect source types and build AI skills using Skill Seekers. Use when the user wants to create skills from documentation, repos, PDFs, videos, or other knowledge sources.
offensive-advanced-redteam
Comprehensive red team operations methodology covering full engagement lifecycle from planning through reporting. Addresses engagement scoping and rules of engagement negotiation, multi-tier C2 infrastructure design with redirectors and domain fronting, malleable traffic profiles and beacon tradecraft, OPSEC…
offensive-crypto-attacks
Systematic methodology for identifying and exploiting cryptographic implementation weaknesses in real-world applications. Covers padding oracle attacks against CBC-mode ciphers with PKCS7 padding (Vaudenay's original attack through modern padbuster automation), ECB mode exploitation including block cut-and-paste and…
offensive-c2-frameworks
Command and Control framework deployment, configuration, and operational tradecraft for red team engagements. Covers Cobalt Strike (malleable C2 profiles, Beacon types HTTP/HTTPS/DNS/SMB, Beacon Object Files for in-memory execution, sleep and jitter tuning, named pipe pivoting), Sliver (implant generation across…