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 lineargit 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/linear)<a href="https://agentmods.dev/skills/asgeirtj/system_prompts_leaks/linear"><img src="https://agentmods.dev/badge/skills/asgeirtj/system_prompts_leaks/linear/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/linear"><img src="https://agentmods.dev/badge/skills/asgeirtj/system_prompts_leaks/linear.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 | $0.00053 | $0.00575 |
| Opus 5.5 | $0.00021 | $0.00230 |
| Sonnet 5.5 | $0.00011 | $0.00115 |
| Haiku 4.5 | $0.00005 | $0.00057 |
Grade A, and why
linear 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 19d 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
Linear
Use the installed linear CLI. Start with linear status. If it reports
not_connected, run linear authorize-url and share only the returned
connect_url with the user.
OAuth uses Dynamic Client Registration and PKCE through authd. No shared client
secret enters Muse, and credentials must never be requested in chat.
The first connection requests read-only access. Before a write, or after one
fails for missing access, run linear status --for-command <permission>. For
Linear's save_* tools, use the matching *.create permission when the
arguments omit id, and the matching *.manage permission when they include
id. If the status command
returns scope_status: not_granted, copy scope_add_url exactly and post it on
its own line as [Additional Linear access](<scope_add_url>) so the client
renders the native access button, then wait for consent to finish before
retrying once. OAuth access does not replace Hatch approval. Never construct a
scope URL or ask for tokens in chat.
Run linear list-tools to inspect the live provider catalogue and schemas,
then call an advertised tool with:
linear call-tool --name <tool> --arguments-json '<json-object>'
list-tools exposes only reviewed Linear tools and includes each tool's
hatch_permission, hatch_action, and hatch_permission_label. For a
polymorphic save_* tool, the permission fields show the id-based selector
used at execution. Linear hides write tools from its live catalogue while the
OAuth grant is read-only; do not describe an all-read tools list as lack of
write support. In that state, list-tools returns manifest-derived
additional_access entries separately from the provider-advertised tools. When
the user requests one of those capabilities, share its scope_add_url using
the exact Additional Linear access link format above, wait for consent, then
rerun linear list-tools to obtain the provider's live write-tool schema.
Unknown or newly advertised provider tools remain unavailable until reviewed.
Read permissions follow the user's connector settings; changes require the
corresponding granular approval. Do not retry a failed or timed-out write
automatically because its side effect may have completed.
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.
- 19d ago First seen · 52 lines · 53 tokens per session scan A 0708c8260bf0
linear is a skill published in the GitHub repository asgeirtj/system_prompts_leaks (70,113 stars, last pushed yesterday), licensed CC0-1.0. It adds 53 tokens to every session and 575 once invoked, about $0.0002 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
kanban-video-orchestrator
Plan and run multi-agent video production pipelines.
sdlc-review
Review Kanban handoffs and route verified outcomes.
teams-meeting-pipeline
Teams meeting summaries, job replay, Graph subscriptions.
meeting-action-items
Turn meeting notes into cited decisions, owners, tickets.
subagent-dag-orchestration
Use when delegated work breaks into several tasks that agents on the roster can carry out. The host's role decides what to delegate; this skill decides how to group that work. A DAG node dispatches to a roster agent, including the generic agent, and cannot call the host's tools. Choose from the tool's subagent…
service-sop
A standard operating procedure for online domestic package-tour enquiries, from collecting trip needs through quoting, availability checks, presentation delivery, and follow-up. It also covers complaints, unusual requests, and group enquiries.