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 google-docsgit 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/google-docs)<a href="https://agentmods.dev/skills/asgeirtj/system_prompts_leaks/google-docs"><img src="https://agentmods.dev/badge/skills/asgeirtj/system_prompts_leaks/google-docs/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/google-docs"><img src="https://agentmods.dev/badge/skills/asgeirtj/system_prompts_leaks/google-docs.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 | $0.00014 | $0.01089 |
| Opus 5.5 | $0.00006 | $0.00436 |
| Sonnet 5.5 | $0.00003 | $0.00218 |
| Haiku 4.5 | $0.00001 | $0.00109 |
Grade A, and why
google_docs 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.
How it starts
The opening of the file, as written. The whole thing — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Google Docs
Purpose
Manage Google Docs through hatch_gws_cli; the vendored Google Workspace CLI is the only implementation path.
Tooling
Use exec to run:
hatch_gws_cli docs <resource> <method> [flags]
Connection management
hatch_gws_cli docs status
hatch_gws_cli docs disconnect
Docs operations
Core patterns:
hatch_gws_cli docs statushatch_gws_cli docs disconnecthatch_gws_cli schema docs.documents.gethatch_gws_cli schema docs.documents.createhatch_gws_cli schema docs.documents.batchUpdate
Common raw API calls:
hatch_gws_cli docs documents get --params '{"documentId":"<document_id>"}'hatch_gws_cli docs documents create --json '{"title":"Project Brief"}'hatch_gws_cli docs documents batchUpdate --params '{"documentId":"<document_id>"}' --json '{"requests":[{"insertText":{"location":{"index":1},"text":"Hello, world!"}}]}'
Vendored helpers are also available:
hatch_gws_cli docs +write --document <document_id> --text 'Hello, world!'
JSON output contract:
status: parsestatus,connect_url, anddisconnect_urldisconnect: parseok,action,status, anddisconnect_url
Composing document content
Never compose a document body through batchUpdate inserts or docs +write, whether new, rewritten, or reconstructed. Inserted plain text arrives unformatted. Build it as a document artifact first. The artifacts tool renders a styled .docx under ~/workspace/your_files/<artifact-slug>/. Then put it in Google:
- Mint the empty doc with
hatch_gws_cli docs documents create --json '{"title":"<title>"}'. - Fill it from the built file:
hatch_gws_cli drive files update --params '{"fileId":"<document_id>"}' --upload "$JARVIS_HOME/workspace/your_files/<artifact-slug>/<name>.docx". Drive converts the upload into the Google Doc in place, formatting intact. The path must be absolute and inside the home directory. Relative and/tmppaths fail. - Read it back with
documents.getand confirm the styling landed before you call it ready. A successful upload proves Drive converted the file, not that the page reads correctly. - To revise, edit the artifact and re-upload to the same document ID. The artifact is the working copy. The upload replaces the whole body, so it discards any edits made in Google since, including the user's. Read the published copy back first. If it changed, say so and wait for a yes.
What ships with it
1 file 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.
- 12d ago First seen · 74 lines · 14 tokens per session scan A 7a2341cd965a
google_docs is a skill published in the GitHub repository asgeirtj/system_prompts_leaks (69,759 stars, last pushed yesterday), licensed CC0-1.0. It adds 14 tokens to every session and 1,089 once invoked, about $0.0001 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.