Ask-human-mcp – zero-config human-in-loop hatch to stop hallucinations
While building my startup i kept running into the issue where ai agents in cursor create endpoints or code that shouldn't exist, hallucinates strings, or just don't understand the code. ask-human-mcp pauses your agent whenever it’s stuck, logs a question into ask_human.md in your root directory with answer: PENDING, and then resumes as soon as you fill in the correct answer. the pain: your agent screams out an endpoint that never existed it makes confident assumptions and you spend hours debugging false leads the fix: ask-human-mcp gives your agent an escape hatch. when it’s unsure, it calls…
In plain words
Ask-human-mcp is a tool that prevents AI agents in Cursor from creating nonexistent code or making false assumptions. When an agent encounters uncertainty, it pauses and logs a question to ask_human.md with a PENDING status. The developer fills in the correct answer, and the agent resumes work. It requires zero configuration beyond a single line in .cursor/mcp.json and works across macOS, Linux, and Windows. This helps developers avoid hours of debugging false leads caused by AI hallucinations.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
While building my startup i kept running into the issue where ai agents in cursor create endpoints or code that shouldn't exist, hallucinates strings, or just don't understand the code. ask-human-mcp pauses your agent whenever it’s stuck, logs a question into ask_human.md in your root directory with answer: PENDING, and then resumes as soon as you fill in the correct answer. the pain: your agent screams out an endpoint that never existed it makes confident assumptions and you spend hours debugging false leads the fix: ask-human-mcp gives your agent an escape hatch. when it’s unsure, it calls ask_human(), writes a question into ask_human.md, and waits. you swap answer: PENDING for the real answer and it keeps going. some features: - zero config: pip install ask-human-mcp + one line in .cursor/mcp.json → boom, you’re live - cross-platform: works on macOS, Linux, and Windows—no extra servers or webhooks. - markdown Q\&A: agent calls await ask_human(), question lands in ask_human.md with answer: PENDING. you write the answer, agent picks back up - file locking & rotation: prevents corrupt files, limits pending questions, auto-rotates when ask_human.md hits ~50 MB the quickstart pip install ask-human-mcp ask-human-mcp --help add to .cursor/mcp.json and restart: { "mcpServers": { "ask-human": { "command": "ask-human-mcp" } } } now any call like: answer = await ask_human( "which auth endpoint do we use?", "building login form in auth.js" ) creates: ### Q8c4f1e2a ts: 2025-01-15 14:30 q: which auth endpoint do we use? ctx: building login form in auth.js answer: PENDING just replace answer: PENDING with the real endpoint (e.g., `POST /api/v2/auth/login`) and your agent continues. link: github -> https://github.com/Masony817/ask-human-mcp feedback: I'm Mason a 19yo solo-founder at Kallro. Happy to hear any bugs, feature requests, or weird edge cases you uncover - drop a comment or open an issue! buy me a coffee -> coff.ee/masonyarbrough
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 26d ago · cactuscompute.com
Launched alongside, June 2025
the whole month →- IM
Hey HN! Pilots everywhere are required to keep a logbook of all their flying hours, aircraft, airports, and so on. Since I track everything digitally (some people still just use paper logbooks!), I put together some data visualizations and a few 3D globes to show my flying history. This globe is probably my favourite so far: https://jameshard.ing/pilot/globes/all If you’ve got ideas for other graphs or ways to show this kind of data, I’d love to hear them!
Life & fun · 2025 · jameshard.ing
Stunning presentations with AI. No design skills required.
AI · 2025 · chroniclehq.com
Build native mobile apps for iOS and Android without code.
Dev tools · 2025 · bble.io
All your connections, across all your socials.
AI · 2025 · pally.com
- WC
I was the main contributor to workout.lol, an open-source fitness app to easily build a workout routine. The project had traction (1.4k GitHub stars, 95 forks, ~20K visits/month), but was eventually sold due to video licensing hurdles. The new owner stopped maintaining it, and the repo went abandoned. Over the next 9 months, I sent 15 emails to try to save it : no replies. Feature requests & issues were ignored. The community was left with a "broken" tool let's say. I couldn't just let it die So I built the new version from scratch with the same open-source spirit, but a better…
Dev tools · 2025 · github.com