agentleFS
Sign inStart free
tim101010101tim101010101/lazyagent/.agents/AGENTS.md

lazyagent

AGENTS.md1 starsChanged 8 months agoMITView on GitHub

Lazygit-style TUI for managing live AI coding agent sessions via tmux. Rust + ratatui + crossterm.

  • 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 31,542 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

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

4 rules. About 0 are common, 1 is distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

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

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

    All tests use #[test], MockProvider with TestBackend — no external test framework

    line 45
  2. Very specific:

    Prefer anyhow::Result for app-level errors

    line 46
  3. Very specific:

    Use /log-search skill to search/filter logs

    line 62
  4. Distinctive:

    Add async to the event loop without strong reason

    line 66

≈ “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

LazyAgent

Lazygit-style TUI for managing live AI coding agent sessions via tmux. Rust + ratatui + crossterm.

Architecture

src/
  main.rs          — entry, event loop, attach/spawn dispatch, render_main
  app.rs           — App state, key handling, session navigation, search, kill confirm
  event.rs         — crossterm event polling
  session.rs       — SessionManager: poll/spawn/attach/kill via TmuxController
  protocol/
    mod.rs         — types: AgentSession, AgentStatus, SessionKind, SessionSource, ExecPlan
    provider.rs    — Provider trait: detect_status, match_process, exec_plan
  provider/
    mod.rs         — provider registry
    claude.rs      — ClaudeProvider: process matching + pane output status detection
  tmux/
    mod.rs         — TmuxController: discover_sessions, spawn_session, attach/kill
  tui/
    mod.rs         — terminal init/restore
    layout.rs      — AppLayout: 2-col or 3-col horizontal
    sidebar.rs     — session list with status icons, source/project grouping
    detail.rs      — detail panel: KV display of session info
    help.rs        — help bar with context-aware hints
    theme.rs       — color/style constants incl status colors

Key Patterns

  • Provider protocol: trait-based plugin system — detect_status() for pane output, match_process() for discovery
  • Session discovery: tmux list-panes -a -F → match process via providers → capture-pane for status
  • Managed sessions: spawned with la/<provider>/<dir> naming convention
  • Discovered sessions: externally-started agent sessions found in any tmux session
  • Attach flow: suspend TUI → tmux attach-session -t <name> → restore TUI on exit
  • Auto-refresh: sessions polled every 2s via tick counter
  • Kill confirmation: d sets confirm_kill, y/n to confirm/cancel

Code Conventions

  • Rust 2021 edition, no async in main loop (sync event polling, 100ms tick)
  • Immutable ratatui frame rendering, no stateful widgets
  • CJK-aware text truncation
  • All tests use #[test], MockProvider with TestBackend — no external test framework
  • Prefer anyhow::Result for app-level errors

Build & Test

cargo build
cargo test        # 24 tests: e2e, provider, sidebar, tmux, rendering
cargo run         # requires tmux

Logging

  • tracing-based structured logging, default info level (override via LAZYAGENT_LOG env)
  • Log dir: ~/.local/state/lazyagent/ (dirs::state_dir())
  • Files: lazyagent.log.YYYY-MM-DD (all), lazyagent-error.log.YYYY-MM-DD (errors only)
  • Daily rotation, 7-day retention
  • Use /log-search skill to search/filter logs

Do Not

  • Add async to the event loop without strong reason
  • Break Provider trait backward compat without updating all providers
  • Hardcode paths — use dirs crate for home, env vars for config
  • Swallow errors silently — always surface via app.error_message
  • Kill agent sessions on TUI quit — sessions must persist independently

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.

Check your setup, free