# Recall what the owner saw, said or heard on their own computer

> Recall what the owner saw, said or heard on their own computer: 0 companies fully do it, 1 with a limit, 165 carry the tag. Every listing in Local AI with the reason and a source.

- Canonical: https://www.anchorterminal.com/niches/local-ai.screen-and-meeting-recall/
- Markdown: https://www.anchorterminal.com/niches/local-ai.screen-and-meeting-recall/index.md (~1,300 tokens)
- Slim: https://www.anchorterminal.com/niches/local-ai.screen-and-meeting-recall/index.min.md (~230 tokens, same facts, less prose, for token-sensitive contexts)
- JSON: https://www.anchorterminal.com/niches/local-ai.screen-and-meeting-recall/index.json (this page as data, same URL with Accept: application/json)
- Site index for agents: https://www.anchorterminal.com/llms.txt (the key docs and every listing in one file, at most 16 MB, about 4 million tokens: https://www.anchorterminal.com/llms-full.txt)
- API: https://www.anchorterminal.com/api/v1/index.json
- MCP: https://www.anchorterminal.com/mcp (the whole catalogue as tools, no key)
- DNS: dig +short TXT <slug>.grade.anchorterminal.com (the grade as a DNS TXT record, for a sandbox with no HTTP where names still resolve; also top., best.<category>. and <a>.vs.<b>. under the zone, and the same under grade.letme.dev)
- Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data)
- Updated: 2026-11-15

No company's docs show a full fit yet. The listings below do the job with a limit, or not at all.

| Companies, full fit | With a limit | Carry the tag | Share of the category's work | First-hand accounts |
| --- | --- | --- | --- | --- |
| 0 | 1 | 165 | 5 | 2 |

## What counts as done

A search over a named time range returns the matching screen text or transcript, and the owner can see what, if anything, left the machine for transcription.

Measured, for our own tests: the search returns within 5 seconds, and a meeting of 3 hours or more is transcribed without gaps.

How agents ask for it: what was I working on before lunch; summarise this morning's Teams call with action items; find the page I had open yesterday about X; give Claude memory of my screen

## Does the job, with a limit

It does the job, but something in the docs or a user's account will get in an agent's way. The limit is named in each row.

- **screenpipe (Negentropy Labs, Inc. (dba Screenpipe))**: search-content takes start_time and end_time and filters by app, speaker and type. A five-hour Teams call was force-ended at two hours with audio dropped, and Free history reads are capped at 24 hours in source.

- not this job: AnythingLLM (Mintplex Labs, Inc.): Not a recording or transcription tool.
- not this job: Berth (Trevin Tindall (T-Crypt on GitHub)): Not a recording tool.
- not this job: BitRun (Abhishek Khanra (IshekKhal on GitHub)): Not a recording tool.
- not this job: Docker Model Runner (Docker, Inc.): Not a recording tool.
- not this job: Foundry Local (Microsoft Corporation): Not a recording tool; Whisper serves per-request transcription.
- not this job: Core (ZMJ, Inc., doing business as Ghost, 325 9th Street, San Francisco, CA 94103, United States): Not a screen or meeting recorder.
- not this job: GPT4All (Nomic, Inc.): Not a recording or transcription tool.
- not this job: HyperQwen (syv.ai (Copenhagen, CVR DK44671247)): Not a recording tool.
- not this job: Jan (Menlo Research Pte Ltd): Not a recording or transcription tool.
- not this job: Khoj (Khoj Inc.): Not a recording or transcription tool.
- not this job: KoboldCpp (LostRuins (Concedo)): Not a recording tool; Whisper serves one-off transcription only.
- not this job: Lemonade (Advanced Micro Devices, Inc.): Not a recording tool.
- not this job: llama.cpp (ggml.ai, part of Hugging Face since 2026): Not a recording or transcription tool.
- not this job: LM Studio (Element Labs, Inc.): Not a recording or transcription tool.
- not this job: LocalAI (Ettore Di Giacinto and the LocalAI team): Not a recording tool; transcription and diarisation are per-request routes.
- not this job: LocalGhost (LocalGhost.ai Ltd (company number 17213100, registered in England and Wales, incorporated 12 May 2026)): Records voice notes and photos from the phone, not the computer's screen or meetings.
- not this job: MLX LM (Apple Inc.): Not a recording tool.
- not this job: Ollama (Ollama Inc.): Not a recording or transcription tool.
- not this job: Open WebUI (Open WebUI Inc.): Not a recording or transcription tool.
- not this job: PrivateGPT (PriBAI Technology Spain S.L.): Not a recording tool.
- not this job: TextGen (oobabooga): Not a recording tool.
- not this job: Underdog (Conway Research): Not a screen or meeting recorder; speech-to-text is for the assistant's own input.
- not this job: vLLM (The Linux Foundation (vLLM is a PyTorch Foundation project)): Not a recording tool.

## What stops agents finishing

- live session force-ended at two hours with audio silently dropped until the call is re-detected
- free plan returns only the last 24 hours
- 33 MCP tools at 5,600 to 8,300 tokens of definitions with no way to pick a subset
- cloud transcription returns 503 cost_control_unavailable and the local fallback is deferred

## Around it

Before: speech-to-text.meeting-with-speakers. After: agent-memory.user-facts-recall, personal-assistants.morning-brief.

## Where this comes from

Weight 5 of 100 in Local AI: 2 of 50 accounts, both screenpipe; the only listing in the category that does this and a small share of what people wrote about

As data: https://www.anchorterminal.com/api/v1/use-cases/local-ai.screen-and-meeting-recall.json, or the MCP tool get_use_case with id local-ai.screen-and-meeting-recall. Draft, read 2026-11-12 from public docs and first-hand accounts, not yet reviewed by a person.

---

Query this catalogue directly over MCP: https://www.anchorterminal.com/mcp (Streamable HTTP, no key). How: https://www.anchorterminal.com/agents/#mcp
