agentleFS
vinhnxvinhnx/VTCode/src/AGENTS.md

VTCode / src

AGENTS.md903 starsChanged 27 days agoApache-2.0View on GitHub

Root AGENTS.md | CLI entrypoint, session bootstrap, agent runloop wiring, and the headless desktop-server JSON-RPC service, whose stdout is protocol-only with internal tags stripped. main.rs entry | cli/ command handlers, WebMCP serving, companion exec supervision | startup/ onboarding and StartupPolicy | updater/ downloads and self-replacement | main_helpers/ tracing and runtime init | codex_app_server/ bridge | allocator.rs, process_hardening.rs process setup agent/runloop/ single-agent loop: unified/turn/session_loop_runner/ facade with orchestration/ owning the loop body, setup, and teardown; unified/turn/turn_loop/ request-boundary compaction and ordered finalization; unified/session_setup/hook_approval.rs…

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 1

    1. vinhnx/vtcode★ 903100% same

    Edited · 0

    None found.

    Wrote their own version · 0

    None found.

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • 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 31,700 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 repository4 other files it gives its agents.

The file

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

2 rules. About 0 are common, 0 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

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

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

    Keep the binary thin; runtime logic, spool preview generation, and shell activity classification belong in vtcode-core, and the binary only …

    line 14
  2. Very specific:

    Use the shared ThreadEvent contract for telemetry.

    line 15

≈ “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
<!-- markdownlint-disable MD013 -->

vtcode (binary)

Root AGENTS.md | CLI entrypoint, session bootstrap, agent runloop wiring, and the headless desktop-server JSON-RPC service, whose stdout is protocol-only with internal tags stripped.

Modules

main.rs entry | cli/ command handlers, WebMCP serving, companion exec supervision | startup/ onboarding and StartupPolicy | updater/ downloads and self-replacement | main_helpers/ tracing and runtime init | codex_app_server/ bridge | allocator.rs, process_hardening.rs process setup agent/runloop/ single-agent loop: unified/turn/session_loop_runner/ facade with orchestration/ owning the loop body, setup, and teardown; unified/turn/turn_loop/ request-boundary compaction and ordered finalization; unified/session_setup/hook_approval.rs workspace hook approval overlay agent/runloop/tool_output/ tracker rows, dispatch, and ANSI diff rendering | agent/runloop/unified/turn/compaction/ delegates to vtcode-core::compaction | agent/runloop/unified/webmcp.rs active WebMCP bridge | agent/runloop/unified/instruction_audit.rs read-only /doctor prompt-audit

Rules

  • Keep the binary thin; runtime logic, spool preview generation, and shell activity classification belong in vtcode-core, and the binary only serializes typed references.
  • Use the shared ThreadEvent contract for telemetry; unexpected exits must emit terminal lifecycle events before draining, and take_active_turn is the single take point for terminal emission.
  • Build LifecycleHookEngine with new_with_session_gated so workspace-controlled hooks stay gated behind the approval overlay.
  • Verification gates completion only, never editing; verifier responses require a terminal exit code before clearing the gate, and verification credit is never derived from request-only projections.
  • Request context is persisted once at a user-turn boundary and never rewritten, and the tool-free recovery directive is request-only so the cached prompt prefix stays byte-identical.
  • Exit is session-terminal: check stop state before recovery, tool admission, and autonomous follow-ups.
  • Updater asset URLs stay on HTTPS GitHub release paths, downloads never use API credentials, and missing or invalid checksums abort installation.
  • Companion setup preserves operator CLI/config policy, never grants trust or inherits proxy credentials, and keeps raw stderr private.
  • Internal harness follow-ups use try_queue_synthetic_follow_up_input and preserve recovery budgets; never classify them as fresh user requests.

Gotchas

  • Session load probes the exclusive event-log lock first and fails fast when another process holds the session.
  • Desktop-server startup is auth-tolerant: missing credentials are recorded in provider_auth_error, never fatal.
  • Async checkpointing acknowledges consumed steering intents only after a Persisted history result, never after a throttled checkpoint.
  • A turn still active 5s after cancel escalates to force-stop and restart from history, and unconfirmed termination holds session_transitioning instead of reporting a stop.
  • session/recent_prompts serves Everywhere/Project scopes from archives via clean_archived_prompts; Session answers empty since the client's live thread owns it.

Details: maintainer notes — read the section for the subsystem you change before editing it.

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.