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 documentgit 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/document)<a href="https://agentmods.dev/skills/asgeirtj/system_prompts_leaks/document"><img src="https://agentmods.dev/badge/skills/asgeirtj/system_prompts_leaks/document/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/document"><img src="https://agentmods.dev/badge/skills/asgeirtj/system_prompts_leaks/document.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 | $0.00124 | $0.00483 |
| Opus 5.5 | $0.00050 | $0.00193 |
| Sonnet 5.5 | $0.00025 | $0.00097 |
| Haiku 4.5 | $0.00012 | $0.00048 |
Grade A, and why
artifact_document 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 14d 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.
What it actually says
Word-document artifacts
A docx is generated with the preinstalled python-docx library from a
generator script. Keep the generator under .src/: it is the editable
source for future revisions, and the binary is always regenerated from it.
| Task | Read first |
|---|---|
| Design and structure | /opt/hatch/skills/artifacts/document/references/visual.md |
| What the words say: outline, headings, tone, the prose read-back | /opt/hatch/skills/artifacts/references/prose.md (shared) |
| Content formatting (tables, lists, emphasis) | /opt/hatch/skills/artifacts/references/markdown.md (shared) |
| The document shows a place or a map | /opt/hatch/skills/artifacts/references/maps.md (shared) |
| Edit an existing or uploaded .docx/.dotx, tracked changes, comments, extract/read content, legacy .doc | /opt/hatch/skills/artifacts/document/references/editing.md |
Set a non-empty document.core_properties.title, use a human-readable
filename and visible title, and never fake structure: real numbering for
lists (never a literal bullet character), real heading styles for anything a
table of contents must see, a paragraph bottom border for a rule (never a
one-row table), and separate paragraphs instead of newlines inside a run.
Verification
Follow /opt/hatch/skills/artifacts/testing/SKILL.md: render the document to PDF with
headless LibreOffice, rasterize with pdftoppm, and read every page image
before returning a link.
What ships with it
2 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.
- 14d ago First seen · 32 lines · 124 tokens per session scan A 7628749f20fa
artifact_document is a skill published in the GitHub repository asgeirtj/system_prompts_leaks (69,863 stars, last pushed 2d ago), licensed CC0-1.0. It adds 124 tokens to every session and 483 once invoked, about $0.0005 per session on Opus 5.5. 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 2026-10-05.
Other skills, from other repositories
dcf-model
Build discounted cash flow valuation workbooks in Excel.
comps-analysis
Build comparable-company valuation workbooks in Excel.
lbo-model
Build leveraged buyout workbooks with IRR/MOIC in Excel.
google-workspace
Gmail, Calendar, Drive, Docs, Sheets via gws CLI or Python.
PDF files: create, read, merge, fill, OCR, edit text.
siyuan
Query and edit a SiYuan knowledge base via its API.