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 gmailgit 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/gmail)<a href="https://agentmods.dev/skills/asgeirtj/system_prompts_leaks/gmail"><img src="https://agentmods.dev/badge/skills/asgeirtj/system_prompts_leaks/gmail/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/gmail"><img src="https://agentmods.dev/badge/skills/asgeirtj/system_prompts_leaks/gmail.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 | $0.00033 | $0.03752 |
| Opus 5.5 | $0.00013 | $0.01501 |
| Sonnet 5.5 | $0.00007 | $0.00750 |
| Haiku 4.5 | $0.00003 | $0.00375 |
Grade A, and why
gmail 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 17d 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 — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Gmail
Everything runs as hatch_gws_cli gmail <command>. The flows below give the exact command for each task, so start there. Commands come in two kinds:
+shortcut (+send,+read,+unsubscribe): the simple, preferred form.- Raw API call: reach any Gmail API method by writing its dotted name as separate words, so
users.messages.listbecomesusers messages list. Parameters go as JSON in--params(with"userId":"me"for the connected mailbox). A write's content goes in--json, the request body.
Raw calls reach the whole API beyond the shortcuts: drafts, labels, threads, history, and settings like the vacation responder, forwarding, and send-as. To find a method, drill --help. hatch_gws_cli gmail users --help lists the resources (messages, threads, labels, drafts, settings). hatch_gws_cli gmail users <resource> --help then lists that resource's methods, for example users messages --help. Then hatch_gws_cli schema gmail.<method> (for example gmail.users.settings.updateVacation) gives that method's --params and --json.
Do not offer or attempt to create, edit, or delete Gmail filters.
Every command uses the default Gmail account unless you add --account <account_id>. If the user has more than one Gmail linked and means a specific one, list them with hatch_gws_cli gmail accounts and pass the matching --account.
Connecting
Gmail needs a one-time connect before commands return data. Run hatch_gws_cli gmail status. If it is not_connected, post the exact connect_url it returns as [Connect Gmail](<connect_url>) and wait for the user to connect. Don't invent a URL, send the user to Settings, or ask for credentials.
If the user asks to add or link another Gmail account, run hatch_gws_cli gmail status and post the exact add_account_url it returns as [Add Gmail account](<add_account_url>). Never reuse connect_url for an additional account. If add_account_url is absent, say that adding another account is unavailable. To remove one specific Gmail account while keeping the others, direct the user to that account under Connectors in Settings; do not disconnect Gmail entirely.
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.
- 17d ago First seen · 129 lines · 33 tokens per session scan A 1a6a0fac1e63
gmail is a skill published in the GitHub repository asgeirtj/system_prompts_leaks (70,001 stars, last pushed 2d ago), licensed CC0-1.0. It adds 33 tokens to every session and 3,752 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
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.