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 ghlgit 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/ghl)<a href="https://agentmods.dev/skills/asgeirtj/system_prompts_leaks/ghl"><img src="https://agentmods.dev/badge/skills/asgeirtj/system_prompts_leaks/ghl/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/ghl"><img src="https://agentmods.dev/badge/skills/asgeirtj/system_prompts_leaks/ghl.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 | $0.00020 | $0.00836 |
| Opus 5.5 | $0.00008 | $0.00334 |
| Sonnet 5.5 | $0.00004 | $0.00167 |
| Haiku 4.5 | $0.00002 | $0.00084 |
Grade A, and why
ghl 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 — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
HighLevel
Always call the product HighLevel in user-facing responses. Never prefix
the name with "Go," and never expand the internal ghl identifier into a
display name.
Use /opt/hatch/bin/ghl status to check the connection. If disconnected, run
/opt/hatch/bin/ghl authorize-url and share the returned connect_url.
The user signs in through HighLevel; never request credentials in chat.
Find the operation and sub-account
Run /opt/hatch/bin/ghl list-operations for the pinned operation catalog and
its granular permission methods. The catalog covers HighLevel's CRM,
communications, scheduling, commerce, content, administration, and AI domains.
CRM/calendar reads and conversation/message reads have separate permissions. Contact edits, notes/tasks, opportunities, appointments, automations, messaging, and CRM deletion each have their own control. A denied message-read permission does not prevent authorized CRM/calendar reads.
Use /opt/hatch/bin/ghl call-tool --name list_locations --arguments-json '{"query":"<sub-account name>"}' to resolve the intended sub-account.
If several match, ask which one the user means. Retain its ID for execution.
Use call-tool --name describe_operation --arguments-json '{"operationId":"<operation ID>"}' for exact path/query/body fields and
required inputs. search_operations searches HighLevel's live catalog. Use an
operation's pinned permission from list-operations when it is present. A
future provider operation that has not reached the pinned catalog still uses a
mandatory one-shot write approval until its permission is reviewed.
list-tools returns live discovery tool schemas. These calls use
/opt/hatch/bin/ghl too.
Read or change data
/opt/hatch/bin/ghl execute-operation \
--operation-id <operation ID> --location-id <sub-account ID> \
--params-json '{"path":{},"query":{},"body":{}}'
Supply path IDs and payload fields from describe_operation. Contact
create/update/upsert cannot include tags; use add-tags or remove-tags.
Opportunity upsert cannot include follower-management fields; use the
separate add/remove follower operations. Only grouped
path, query, and body objects are accepted; omit empty groups as needed.
For each logical write/delete, also supply a fresh --idempotency-key <UUID>.
An optional --reason explains the action. --dry-run previews the resolved
request without changing CRM data and does not require an idempotency key.
A dry run does not authorize the subsequent write.
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.
- 12d ago First seen · 80 lines · 20 tokens per session scan A 36d56c53e5fb
ghl is a skill published in the GitHub repository asgeirtj/system_prompts_leaks (69,759 stars, last pushed today), licensed CC0-1.0. It adds 20 tokens to every session and 836 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
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 a task breaks into several distinct steps that a separate sub-agent could each carry out. A DAG node dispatches to an agent on the roster and cannot call your own tools, so steps that are just your own tool calls (reading files, small edits, running commands) are not a DAG for being many - do those yourself.…
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.