agentleFS
Zackriya-SolutionsZackriya-Solutions/meeting-minutes/AGENTS.md

meeting-minutes

AGENTS.md33k starsChanged 3 days agoMITView on GitHub

These instructions apply to AI coding agents working in this repository. See architecture and audio pipeline, including frontend/native communication and model management. Use pnpm 9.15.9. Run these commands from frontend/: Before running desktop dev/build commands, complete the first-run sidecar setup. Installing frontend dependencies alone does not stage llama-helper. Use the Tauri app for development and validation; a standalone Next.js server does not exercise native commands, audio, or permissions. See build and platform setup, development commands and debugging, and GPU acceleration…

  • Installs packages

Trending

Quiet since September.about the whole repository

Dec 2025: 1 mentionJan 2026: 5 mentionsFeb 2026: 1 mentionMar 2026: 0 mentionsApr 2026: 1 mentionMay 2026: 1 mentionJun 2026: 0 mentionsJul 2026: 2 mentionsAug 2026: 0 mentionsSep 2026: 1 mentionOct 2026: 0 mentionsNov 2026: 0 mentions
Dec 2025Jun 2026Nov 2026
Mentions a month of the Zackriya-Solutions/meeting-minutes repository that our weekly crawl found. There is no adoption data for this file yet.

Who's talking about Zackriya-Solutions/meeting-minutes

The biggest 5 of 33 mentions linking to the repository that our crawl found. Each line is their own headline.

See all 33

Talked about on GitHub, Bluesky, YouTube, Hacker News and the web (33 times).

On GitHub · 22

17 more on GitHub

On Bluesky · 5

On the web · 1

  • Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.

    Of the AGENTS.md files we read, the 31,439 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.

  • In this repositoryOne other file it gives its agents.

The file

Show its rules and text

We split this AGENTS.md into the rules it gives an agent

14 rules. About 1 is common, 8 are distinctive, and 5 are very specific to this project.

  • ≈1

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use pnpm 9.15.9. Run these commands from frontend/:

    line 17
  2. Very specific:

    Before running desktop dev/build commands, complete the first-run sidecar setup.

    line 19
  3. Very specific:

    Use the Tauri app for development and validation.

    line 27
  4. Distinctive:

    Read the relevant source before editing and follow existing patterns.

    line 33
  5. Distinctive:

    Check source and configuration before repeating technical examples or assumptions from documentation.

    line 35
  6. Distinctive:

    Follow existing shared-state synchronization and async boundaries.

    line 41
  7. Very specific:

    Use perf_debug!() and perf_trace!() for hot-path logging that should disappear in release builds.

    line 42
  8. Distinctive:

    Use Tauri path APIs for application files.

    line 44
  9. Distinctive:

    Use "microphone" and "system" consistently for audio-device roles.

    line 45
  10. Very specific:

    Use existing frontend tests under frontend/tests/ and Rust tests for the affected subsystem.

    line 68
  11. Common:

    Use synthetic or redacted reproduction data.

    line 71
  12. Distinctive:

    When preparing an issue or PR, read and follow CONTRIBUTING.md for duplicate searches, coordination with existing work, branch targets, …

    line 75
  13. Distinctive:

    Do not commit, push, create issues or PRs, or post comments without explicit user authorization for that action.

    line 76
  14. Distinctive:

    Before requesting publication authorization, provide related issue/PR links, a concise change explanation, verification results, and known …

    line 77

≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
AGENTS.md

Agent Guide for Meetily

These instructions apply to AI coding agents working in this repository.

Project Overview

  • Meetily is a desktop application for recording meetings, transcribing audio, and generating summaries.
  • The supported app lives in frontend/: Next.js/React provides the UI, and the Rust/Tauri core handles native behavior, audio, persistence, transcription, and summary orchestration.
  • Local transcription uses Whisper or Parakeet. Summaries use the configured local or cloud provider; do not assume every provider runs locally.
  • Frontend code invokes Tauri commands; Rust emits events that update frontend state. Preserve existing command contracts, event payloads, and state ownership.
  • backend/ is an unsupported legacy archive, not a runtime dependency or a place to add new application behavior. Its FastAPI/CORS behavior is not a supported production API.

See architecture and audio pipeline, including frontend/native communication and model management.

Development Commands

Use pnpm 9.15.9. Run these commands from frontend/:

Before running desktop dev/build commands, complete the first-run sidecar setup. Installing frontend dependencies alone does not stage llama-helper.

CommandPurpose
pnpm install --frozen-lockfileInstall the committed dependency set
pnpm run tauri:devRun the full desktop app
pnpm run tauri:buildBuild the desktop app for production

Use the Tauri app for development and validation; a standalone Next.js server does not exercise native commands, audio, or permissions.

See build and platform setup, development commands and debugging, and GPU acceleration for prerequisites and platform-specific options. After sidecar preparation, prefer the package scripts over clean-run wrappers, which can remove dependencies and generated output.

Working in This Repository

  • Read the relevant source before editing and follow existing patterns. Keep changes focused on the requested problem.
  • Implement native behavior through existing Rust services and Tauri commands/events. Check callers and listeners when changing an interface.
  • Check source and configuration before repeating technical examples or assumptions from documentation.

Engineering Guardrails

  • Preserve both audio outputs: recordings receive mixed audio, while transcription receives VAD-selected speech. A change to one path must not silently break the other.
  • Preserve sample-rate conversions and the channel-drain/flush behavior during recording shutdown. Read the current pipeline and VAD implementation rather than assuming timing constants.
  • Follow existing shared-state synchronization and async boundaries. Do not block async tasks or audio callbacks with expensive work.
  • Use perf_debug!() and perf_trace!() for hot-path logging that should disappear in release builds. Keep pipeline and worker diagnostics aggregated rather than adding per-chunk logs.
  • Reuse AudioBufferPool where applicable instead of adding avoidable audio-buffer allocations.
  • Use Tauri path APIs for application files; do not hardcode platform-specific paths. Preserve platform permission handling.
  • Use "microphone" and "system" consistently for audio-device roles. Keep errors user-friendly and logs diagnostic without exposing private meeting content.

Source Map

AreaStart here
Tauri entry point and command registrationlib.rs
Meeting persistencedatabase/
Recording commands and lifecyclerecording_commands.rs, recording_manager.rs
Device discovery and platform configurationdevices/
Audio permissionspermissions.rs
Audio capturecapture/
Mixing, resampling, and VADpipeline.rs, vad.rs
Recording outputrecording_saver.rs
Transcription providers and workerstranscription/
Whisper model loading and accelerationwhisper_engine/
Parakeet model loading and inferenceparakeet_engine/
Summary providers and orchestrationsummary/
Recording UI and shared frontend statepage.tsx, SidebarProvider.tsx, RecordingStateContext.tsx

Verification and Privacy

  • Exercise the changed path in the Tauri app when application behavior changes. Check relevant platform permissions, devices, and models for native/audio work.
  • Use existing frontend tests under frontend/tests/ and Rust tests for the affected subsystem. Inspect their runner and configuration before choosing commands; do not assume a pnpm test script exists.
  • For documentation-only changes, check links and any changed commands or configuration rather than building the app unnecessarily.
  • Report exactly what ran, observed results, the platform, and what remains untested. Never invent build, test, or hardware results.
  • Use synthetic or redacted reproduction data. Never share private recordings, transcripts, credentials, or unredacted logs and attachments.

Issues, PRs, and Authorization

  • When preparing an issue or PR, read and follow CONTRIBUTING.md for duplicate searches, coordination with existing work, branch targets, review expectations, and submission requirements. Local-only edits and investigations do not require a GitHub duplicate search.
  • Do not commit, push, create issues or PRs, or post comments without explicit user authorization for that action. Permission to edit locally is not permission to publish.
  • Before requesting publication authorization, provide related issue/PR links, a concise change explanation, verification results, and known limitations.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.