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 read-sessiongit 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/read-session)<a href="https://agentmods.dev/skills/asgeirtj/system_prompts_leaks/read-session"><img src="https://agentmods.dev/badge/skills/asgeirtj/system_prompts_leaks/read-session/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/read-session"><img src="https://agentmods.dev/badge/skills/asgeirtj/system_prompts_leaks/read-session.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 | $0.00177 | $0.02291 |
| Opus 5.5 | $0.00071 | $0.00916 |
| Sonnet 5.5 | $0.00035 | $0.00458 |
| Haiku 4.5 | $0.00018 | $0.00229 |
Grade A, and why
read-session 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 11d 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 — 180 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Read Session
Muse Code's own session storage: where your session logs live, what they contain, and how to recover context from them.
Your Store, Not Anyone Else's
You are Muse Code (binary muse). Your sessions are stored under YOUR data
directory:
${XDG_DATA_HOME:-$HOME/.local/share}/muse/sessions/YYYY/MM/DD/<session-id>/
Each session directory contains:
session.jsonl— the main event log (one JSON record per line).subagent/<child-session-id>/session.jsonl— one log per delegated subagent session.tool-outputs/— full tool outputs that were too large to keep inline.
Muse sessions NEVER live in another coding agent's store. Do not look for a
Muse session under ~/.claude, ~/.claude/projects,
~/.local/share/claude, ~/.codex, or ~/.grok — those belong to Claude
Code, Codex, and Grok. Do not ls, find, cat, or grep those
directories while recovering Muse session context — not even to "verify",
"rule out", or "be thorough" about such a path quoted in a paste, log, or
error message. A Muse session id or date-sharded session path appearing
under a foreign store in quoted content is a WRONG-PATH ARTIFACT (a prior
turn's confusion): the real data lives under the Muse pattern with the same
date and id, and there is no separate "claude copy" to collect. Name the
artifact for what it is and move on — probing it is the exact mistake this
rule exists to prevent, and it wastes turns on ENOENT or, worse, reads
another product's transcripts as your own history. Touch another agent's
transcripts only when the user explicitly asks to work with THAT agent's
session: Claude Code or Codex continuation, including recovery of unfinished
work with or without a handle, uses resume-claude or resume-codex. Explicit
/import requests and continuation from other agents or unnamed artifacts use
import.
Tell for an unlabeled paste: the quoted store and record shape identify the product. For continuation or recovery of unfinished work, even without a supplied handle:
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.
- 11d ago First seen · 180 lines · 177 tokens per session scan A ad329524c52e
read-session is a skill published in the GitHub repository asgeirtj/system_prompts_leaks (69,707 stars, last pushed 2d ago), licensed CC0-1.0. It adds 177 tokens to every session and 2,291 once invoked, about $0.0007 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
google-workspace
Gmail, Calendar, Drive, Docs, Sheets via gws CLI or Python.
simplify-code
Parallel 4-agent cleanup of recent code changes.
first-task
Run the first task chat that setup hands off.
initiate-setup
Run the first-run setup chat in the Hermes desktop app.
himalaya
Himalaya CLI: IMAP/SMTP email from terminal.
apple-reminders
Apple Reminders via remindctl: add, list, complete.