System Prompts Leaks is a collection of captured system instructions used to guide AI chatbots and coding agents before they receive user messages. It serves researchers and developers studying how different AI assistants are directed.
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 asgeirtj/system_prompts_leaks --skill claude-api-in-prototypesgit clone --depth 1 https://github.com/asgeirtj/system_prompts_leaksWrote 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/asgeirtj/system_prompts_leaks/claude-api-in-prototypes)<a href="https://agentmods.dev/skills/asgeirtj/system_prompts_leaks/claude-api-in-prototypes"><img src="https://agentmods.dev/badge/skills/asgeirtj/system_prompts_leaks/claude-api-in-prototypes/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/asgeirtj/system_prompts_leaks/claude-api-in-prototypes"><img src="https://agentmods.dev/badge/skills/asgeirtj/system_prompts_leaks/claude-api-in-prototypes.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- Snyk pass
- NVIDIA SkillSpector pass
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 | $0.00019 | $0.00279 |
| Opus 5 | $0.00010 | $0.00139 |
| Sonnet 5 | $0.00004 | $0.00056 |
| Haiku 4.5 | $0.00002 | $0.00028 |
Grade A, and why
claude-api-in-prototypes 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 12d 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- claude-api-in-prototypes — 89% identical, 3 lines differ
What it actually says
Claude API in prototypes
Your HTML artifacts can call Claude via a built-in helper. No SDK or API key needed.
<script>
(async () => {
const text = await window.claude.complete("Summarize this: ...");
// or with a messages array:
const text2 = await window.claude.complete({
messages: [{ role: 'user', content: '...' }],
});
})();
</script>
Calls default to claude-haiku-4-5 with a 1024-token output cap. The body may also set model (haiku/sonnet families only), max_tokens (up to 32000), system, tool_choice, and client tools — standard Messages API shapes, except each tool also carries run: async (input) => string and the helper executes tool calls in-page and loops (max 8 model calls), resolving with the final text. Handler throws become is_error tool_results. Server tools (web search etc.) are rejected; no streaming; rate-limited 15 calls/minute per user, loop iterations included. Shared artifacts run under the viewer's quota.
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.
- 12d ago First seen · 24 lines · 19 tokens per session scan A 7875558b123c
claude-api-in-prototypes is a skill published in the GitHub repository asgeirtj/system_prompts_leaks (64,614 stars, last pushed 2d ago), licensed CC0-1.0. It adds 19 tokens to every session and 279 once invoked, about $0.0001 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
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.
darwinian-evolver
Evolve prompts/regex/SQL/code with Imbue's evolution loop.
tanstack-ai
Build AI features in a TypeScript app with TanStack AI (@tanstack/ai): chat and streaming, tool calling and agents, structured outputs, MCP, code mode, sandboxes and coding-agent harnesses, persistence and resumable streams, memory, embeddings, image, video, and voice generation, and the useChat hook for React, Vue…