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 agentmods add skills/thejefflarson/soundcheck/model-theftnpx skills add thejefflarson/soundcheck --skill model-theftgit clone --depth 1 https://github.com/thejefflarson/soundcheckWhat 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 | $0.00047 | $0.00676 |
| Opus 5 | $0.00023 | $0.00338 |
| Sonnet 5 | $0.00009 | $0.00135 |
| Haiku 4.5 | $0.00005 | $0.00068 |
Grade A, and why
model-theft 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 2d 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 — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Model Theft (OWASP LLM10:2025)
What this checks
Prevents unauthorized replication of proprietary models through API abuse. Unauthenticated or unthrottled inference endpoints let attackers systematically query a model to reconstruct its weights or distill a clone — stealing the commercial and IP value of the deployment.
Vulnerable patterns
- Inference endpoint has no authentication — any client can query freely
- Rate limiting applied per IP only, trivially bypassed with rotating proxies
- Response includes raw
logprobsor full embedding vectors, enabling extraction - No monitoring for systematic/grid-search query patterns that signal extraction attempts
Fix immediately
Flag the vulnerable code and explain the risk. Then suggest a fix that establishes these properties:
- Every inference endpoint requires authentication — API key, bearer token, or mTLS. Unauthenticated endpoints are free training data for anyone who wants to clone the model.
- Rate limits are keyed on the authenticated principal, not the IP. IP-only throttles are defeated by rotating proxies and residential IP pools; a per-user or per-key quota follows the attacker even as IPs churn.
- Extraction-signal fields are stripped from responses. Log-probabilities, full embedding vectors, and per-token probabilities are the primary signals distillation attacks use to reconstruct a model. If a caller does not strictly need them, do not return them.
- Query patterns are monitored for extraction signatures — high-volume, low-entropy, systematic grid-search probes. Alerts fire on anomalies; the handler records user identity, timestamp, and prompt (or a content fingerprint) for after-the-fact investigation.
Translate each principle to the serving framework, auth provider, and rate-limiter of the audited file. Use the framework's documented authentication and throttling middleware — do not roll your own.
Verification
- Every inference endpoint requires a valid API key or bearer token
- Rate limits are enforced per authenticated user, not per IP address
- Log-probabilities, raw embeddings, and weight data are excluded from API responses
- Query logs include user identity, timestamp, and either the prompt itself or a stable fingerprint (hash, embedding, or normalized form) sufficient to detect content-pattern anomalies. Logging only metadata (length, token count, request id) without any reconstructable prompt signal does not satisfy this. Choice between raw prompt and fingerprint is a privacy tradeoff — document the decision.
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.
- 2d ago First seen · 61 lines · 47 tokens per session scan A d2009f33653d
model-theft is a skill published in the GitHub repository thejefflarson/soundcheck (20 stars, last pushed 1mo ago), licensed MIT. It adds 47 tokens to every session and 676 once invoked, about $0.0002 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
make-skill
Use this skill when sedimenting a session into a reusable workspace skill. Triggers when the user wants to turn the current conversation, workflow, or troubleshooting path into a SKILL.md. Phrases like 'turn this into a skill', 'remember how I did X', 'save this workflow', 'make a skill from this', and any /make-skill…
make-skill
用于把当前会话沉淀为可复用的 workspace skill。当用户希望把当前对话、工作流或排错路径写成 SKILL.md 时触发。触发表达包括「把这个变成 skill」「记住我是怎么做 X 的」「保存这个工作流」「make a skill from this」以及任何 /make-skill 调用。.
terraform-skill
Use when working with Terraform or OpenTofu - creating modules, writing tests (native test framework, Terratest), setting up CI/CD pipelines, reviewing configurations, choosing between testing approaches, debugging state issues, implementing security scanning (trivy, checkov), or making infrastructure-as-code…
docx
Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of "Word doc", "word document", ".docx", or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when…
docx
当用户需要创建、读取、编辑或处理 Word 文档（.docx）时，使用此技能。触发场景包括提到“Word 文档”、“.docx”，或要求生成带目录、标题、页码、信头等格式的专业文档；也包括提取或重组 .docx 内容、插入或替换图片、在 Word 文件中查找替换、处理修订或批注，以及将内容整理为正式 Word 文档。如果用户要求生成“报告”“备忘录”“信函”“模板”等 Word / .docx 交付物，也应使用此技能。不要用于 PDF、电子表格、Google Docs，或与文档生成无关的一般编程任务。.
multi_agent_collaboration
Use this skill when another agent's expertise or context is needed, or when the user explicitly asks to involve another agent. First list agents, then use qwenpaw agents chat for two-way communication with replies.