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 ZenlixAI/hitl-mcp --skill hitlgit clone --depth 1 https://github.com/ZenlixAI/hitl-mcpWrote 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/zenlixai/hitl-mcp/hitl)<a href="https://agentmods.dev/skills/zenlixai/hitl-mcp/hitl"><img src="https://agentmods.dev/badge/skills/zenlixai/hitl-mcp/hitl/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/zenlixai/hitl-mcp/hitl"><img src="https://agentmods.dev/badge/skills/zenlixai/hitl-mcp/hitl.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 | $0.00039 | $0.01251 |
| Opus 5 | $0.00019 | $0.00626 |
| Sonnet 5 | $0.00008 | $0.00250 |
| Haiku 4.5 | $0.00004 | $0.00125 |
Grade A, and why
hitl 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 — 179 lines — stays where its author put it; the contents beside it link to each section on GitHub.
hitl
Overview
Use hitl-mcp whenever the agent needs a human answer, approval, clarification, or confirmation.
Non-Negotiable Rule
hitl_ask is never the end of the workflow.
After every hitl_ask, the agent MUST immediately call hitl_wait.
Do not pause to reason about whether waiting is necessary. Do not treat ask as a notification-only action. Do not leave the question pending for "later" in the same agent flow. If the agent chose to ask, the next tool call must be hitl_wait.
The current public model is question-only:
- ask questions
- read pending questions in caller scope
- submit partial answers or skips
- wait on scope progress immediately after asking
Do not model public workflows around request IDs, groups, or interactions.
Critical rule: after every hitl_ask, call hitl_wait immediately. Asking alone only creates pending questions; it does not consume human feedback.
Caller Scope
Caller scope is transport-derived:
agent_identityagent_session_id
The server uses caller scope to isolate pending questions and wait behavior.
MCP Workflow
- Call
hitl_ask. - Immediately call
hitl_wait. - Inspect the wait result.
- If progress is still not terminal, call
hitl_waitagain when appropriate. - Use
hitl_get_pending_questionsonly to render or recover pending state, never as a substitute for the first post-ask wait. - Submit newly answered questions with
hitl_submit_answers. - Optionally skip optional questions with
skipped_question_ids. - Use
hitl_cancel_questionsif pending questions should be cancelled.
Required sequence:
hitl_ask -> hitl_wait -> inspect result -> continue or hitl_wait again
Forbidden sequence:
hitl_ask -> stop
Forbidden sequence:
hitl_ask -> do unrelated work -> hitl_wait
Forbidden sequence:
hitl_ask -> hitl_get_pending_questions -> stop
Tool Reference
hitl_ask: create one or more pending questionshitl_wait: wait on current caller scopehitl_get_pending_questions: list all current pending questionshitl_submit_answers: submit newly answered or skipped questionshitl_cancel_questions: cancel some or all pending questionshitl_get_question: fetch one question byquestion_id
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 · 179 lines · 39 tokens per session scan A e5aa4afd5ee5
hitl is a skill published in the GitHub repository ZenlixAI/hitl-mcp (2 stars, last pushed 1mo ago), licensed MIT. It adds 39 tokens to every session and 1,251 once invoked, about $0.0002 per session on Opus 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 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…