agentleFS
standardbeaglestandardbeagle/agnt/.claude/skills/search/SKILL.md

search

Skill22 starsChanged 11 months agoApache-2.0View on GitHub

Navigate the agnt codebase — find where features live, how subsystems connect, and which files own what

56 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use fetch, integrate and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • stericishereSkill

    fetch

    General-purpose router for fetching ANY data from the internet or external sources. Use when the user needs to pull data from Reddit, YouTube, newsletters, NotebookLM, web articles, APIs, GitHub trending, or any external platform. Handles transcripts, threads, articles, notebooks, trends, deep research, market research, and raw web fetching. Also handles NotebookLM content synthesis — mind maps, knowledge hubs, Obsidian Canvas visualizations, and section-level deep summaries (replaces the old notebooklm-synthesizer skill). Triggers: 'fetch', 'pull from', 'get from', 'download from', 'grab', 'scrape', 'extract from', 'what does Reddit say', 'get the transcript', 'fetch newsletter', 'research this', 'summarize notebooklm', 'create mind map', 'turn into notes', any URL the user wants data from.

    Also in 3 projects that use this

  • modeled-information-formatSkill

    integrate

    This skill should be used when the user asks to "integrate mnemonic", "wire plugin", "add memory to plugin", "enable memory capture in plugin", "integrate memory operations", "add mnemonic protocol", "remove mnemonic integration", "rollback plugin integration", or "migrate legacy memory sections". It wires mnemonic memory capture and recall workflows into other Claude Code plugins using sentinel markers.

    Also in 3 projects that use this

  • soilmassSkill

    export

    Animation export configurations including format selection, resolution, framerate, quality, and optimization settings. Use when rendering animations for web, mobile, video, or social media. Provides tokens for all export parameters across Lottie, GIF, video, and sprite sheet formats.

    Also in 4 projects that use this

  • rogue-dev-studioSkill

    monitoring

    Claude monitoring and usage tracking tools (UV, claude-monitor, claude-usage-cli). Use when the project needs this capability or the user / team manifest asks for it. Use for specialized `monitoring` work when listed in TEAM.yaml or explicitly requested.

    Also in 3 projects that use this

  • DockerSkill

    write

    Write or edit reader-facing technical prose for immediate comprehension. Use for documentation, PR titles and descriptions, release notes, design documents, user-facing explanations, and substantive comments. Also use when asked to make writing clearer, more natural, less AI-generated, or easier to read. For documentation fixes, handles edits, formatting, self-review, and commits after research identifies what to change. Do not use for code-only tasks with no prose deliverable.

    Also in 3 projects that use this

  • JaimeGBASkill

    ingest

    Digest an external source (article, paper, transcript, URL, book chapter) and integrate its content into the knowledge base

    Also in 4 projects that use this

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 48 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill when you need to find where something is implemented, understand how subsystems connect, or orient yourself before making …

    line 8
  2. Distinctive:

    Use string parameters or interfaces to cross this boundary

    line 130

“Common” is measured: kept by at least 6 in 10 of the 48 edited versions of this file we could read. “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
SKILL.md

agnt Codebase Search Guide

Use this skill when you need to find where something is implemented, understand how subsystems connect, or orient yourself before making changes.

Package Map

PackageOwns
cmd/agnt/CLI entry points, PTY wrapper, MCP server bootstrap
internal/daemon/Background service, session/process/proxy lifecycle, IPC hub
internal/proxy/Reverse HTTP proxy, traffic logging, JS injection, WebSocket
internal/tools/All MCP tool implementations and schemas
internal/overlay/Terminal status bar, popup menus, ANSI output filtering
internal/chromedp/Browser automation sessions (CDP)
internal/browser/Chrome/Chromium process lifecycle
internal/tunnel/Cloudflare/ngrok tunnel management
internal/snapshot/Visual regression: baseline capture, image diffing
internal/aichannel/AI agent CLI/API abstraction (Claude, Gemini, Copilot…)
internal/automation/Prompt-driven task processing via claude-go
internal/project/Project type detection (Go/Node/Python)
internal/config/.agnt.kdl parsing, proxy/script config
internal/protocol/IPC verb definitions and request/response types
internal/store/Persistent key-value storage (global/folder/page scopes)
internal/debug/Debug logging to ~/.cache/agnt/logs/

Feature Search Index

MCP Tools

  • All tool registrations: grep -r "Register.*Tools\|mcp.NewTool" internal/tools/
  • Tool handler pattern: internal/tools/daemon_tools.go — handleProxy*, handleProc*
  • Tool schemas (input/output structs): grep "json:\"" internal/tools/*.go
  • get_incidents tool: internal/tools/get_incidents.go; the unified-error collectors proc snapshot still uses live in internal/tools/unified_error.go
  • responsive_audit tool: internal/tools/responsive_audit.go
  • daemon MCP tools (proc, proxy, proxylog, daemon): internal/tools/daemon_tools.go

Daemon — Process Lifecycle

  • Process start/stop: internal/daemon/daemon.go — StartScript, autostartScript, CleanupSessionResources
  • Autostart orchestration: daemon.go — RunAutostart, startAutostartScripts
  • Dependency ordering: internal/daemon/ready_signaler.go — ReadySignaler (SignalReady/SignalFailed, WaitReadyCtx, StartPortProbe); pending set in proc_pending.go, daemon_proc_deps.go
    • See also: docs/plans/2026-03-15-process-dependency-ordering-design.md
  • Process state machine: internal/daemon/daemon.go + go-cli-server/process
  • Script registry (ephemeral, rebuilt per session): go-cli-server/script.Registry
    • CleanupSessionResources clears it on last session disconnect
  • Auto-restarter: internal/daemon/auto_restart.go
  • Port conflict pre-flight: internal/daemon/port_preflight.go
  • Alert scanning (compile errors, panics): internal/overlay/alerts.go → internal/daemon/alert_store.go

Daemon — Proxy Lifecycle

  • Event pipeline (URLDetected → proxy creation): internal/daemon/proxy_events.go
    • handleURLDetected — URL in script output → find matching proxy config → create proxy
    • handleFallbackPortCheck — fallback-port used when URL detection fails
    • handleScriptStopped — cleans up proxies linked to a script
    • See also: .claude/rules/proxy-events.md
  • Proxy creation from config: internal/daemon/daemon.go — autostartProxy
  • Script→proxy tracking (which proxies belong to which script): scriptProxies map + trackScriptProxy

Daemon — Sessions & IPC

  • Hub handlers (all IPC verbs): internal/daemon/hub_handlers.go
    • PROC LIST/START/STOP, PROXY LIST/START/STOP, SCRIPT LIST, SESSION LIST, ALERTS QUERY, DOCTOR RUN
  • Session registration: internal/daemon/daemon.go — handleSessionRegister
  • Session registry: go-cli-server/session.Registry
  • IPC verb constants: internal/protocol/commands.go
  • Client-side IPC: internal/daemon/client.go

Daemon — Health & Diagnostics

  • Doctor checks (all check functions): internal/daemon/doctor.go
    • checkProcessHealth, checkProxyHealth, checkMissingProxies, checkRunningWithErrors, checkSessionHealth, checkStartupErrors, checkProcessTree
  • Startup log (per-process events): internal/daemon/startup_log_store.go
  • Alert store (ring buffer of process alerts): internal/daemon/alert_store.go
  • Rogue process detection: hub_handlers.go — detectRogueProcess
  • Reconciliation model: .claude/rules/daemon-architecture.md

Reverse Proxy

  • ProxyServer (HTTP reverse proxy + WS + JS injection): internal/proxy/server.go
  • ProxyManager (registry, fuzzy lookup, path-scoped): internal/proxy/manager.go
    • GetWithPathFilter — project-scoped lookup
    • ErrProxyAmbiguous — multiple fuzzy matches
  • Traffic logging (1000-entry ring buffer, 14 log types): internal/proxy/logger.go
  • JS injection into HTML responses: internal/proxy/injector.go
  • WebSocket upgrade: internal/proxy/websocket.go
  • URL tracking (detects dev server URLs in process output): internal/daemon/urltracker.go
  • URL matchers (config-driven regex): internal/daemon/urlmatcher.go

Overlay & PTY

  • PTY wrapper entry point: cmd/agnt/run.go — runCommand
  • Output filtering (blocks alt-screen, clamps scroll region): internal/overlay/filter.go
  • Output gate (freeze/unfreeze during menus): internal/overlay/gate.go
  • Summarizer (formats process/proxy data for overlay): internal/overlay/summarizer.go
  • Alert scanner (classifies output: compile errors, panics, warnings): internal/overlay/alerts.go
  • Cross-process data flow: cmd/agnt/pty_common.go — wires scanner → daemon alert push

Browser Automation

  • CDP session management: internal/chromedp/manager.go, session.go
  • Screenshot capture: internal/chromedp/session.go — CaptureScreenshot
  • Browser process lifecycle: internal/browser/browser.go

Configuration

  • .agnt.kdl parsing: internal/config/agnt.go
  • Proxy config (script linking, fallback-port, url-matchers): internal/config/agnt.go — ProxyConfig
  • Script config (run, cwd, depends-on, ports, autostart): internal/config/agnt.go — ScriptConfig
  • Dependency parsing + topo sort: internal/config/agnt_deps.go
  • Config contracts (what fields must be honored): .claude/rules/config-contracts.md

Platform (WSL / Windows / Linux)

  • Platform detection: internal/platform/wsl.go — IsWSL(), ShouldUseWindowsShell(path)
  • Never use runtime.GOOS directly in WSL-aware code — always use platform package
  • Cross-platform mandate: .claude/rules/daemon-architecture.md

Key Architectural Rules

Before changing anything in internal/daemon/ read:

  • .claude/rules/daemon-architecture.md — personas, source of truth, silent failure rules
  • .claude/rules/daemon-lifecycle.md — process/proxy state machines, session ownership
  • .claude/rules/proxy-events.md — event pipeline, known silent failure points

Before changing anything in internal/proxy/ or internal/config/ read:

  • .claude/rules/config-contracts.md — fields that are parsed must be acted on

Import cycle constraint:

  • daemon cannot import overlay (overlay imports daemon via status types)
  • Use string parameters or interfaces to cross this boundary
  • Conversion happens in cmd/agnt/pty_common.go

Common Search Patterns

# Find where a specific IPC verb is handled
grep -n "VerbProxy\|\"PROXY\"" internal/daemon/hub_handlers.go internal/protocol/commands.go

# Find all places a ProxyEvent type is used
grep -n "URLDetected\|ScriptStopped\|FallbackPortCheck\|ExplicitStart" internal/daemon/

# Find all MCP tool input/output structs
grep -n "type.*Input\|type.*Output" internal/tools/*.go

# Find all places ProcessReadiness is called (dependency coordination)
grep -n "MarkReady\|MarkFailed\|MarkExited\|MarkStarting\|\.Wait(" internal/daemon/daemon.go

# Find all places the script registry is touched
grep -rn "scriptRegistry\." internal/daemon/daemon.go

# Find where a specific config field is read vs. acted on
grep -rn "FallbackPort\|fallback.port" internal/

# Find all hub handler functions (IPC verb implementations)
grep -n "^func (d \*Daemon) hubHandle" internal/daemon/hub_handlers.go

# Find all doctor check functions
grep -n "^func check" internal/daemon/doctor.go

Data Flow Diagrams

Dev Server URL → Proxy Creation

Process stdout → URLTracker.Scan()
  → url-matchers (or regex scan)
  → onURLDetected(processID, url)
    → processReadiness.MarkReady(processID)
    → proxyEvents <- URLDetected{...}
      → handleURLDetected()
        → load .agnt.kdl
        → find ProxyConfig where Script == scriptName
        → proxym.Create(proxyID, targetURL)

Process Exit → Dependency Unblocked

ProcessManager removes process
  → URLTracker.cleanupRemovedProcesses()
    → onProcessStopped(processID)
      → processReadiness.MarkExited(processID)  ← unblocks all waiters immediately
      → proxyEvents <- ScriptStopped{...}
        → handleScriptStopped() → stop linked proxies

MCP Tool Call → Daemon IPC

AI Agent calls MCP tool (e.g. proc {action:"list"})
  → internal/tools/daemon_tools.go handler
    → DaemonTools.client.SendCommand("PROC LIST", filter)
      → Unix socket → daemon hub_handlers.go
        → hubHandleProcList() → ProcessManager.List()
          → JSON response → tool output

Session Connect → Autostart

agnt run / Claude Code session
  → daemon.handleSessionRegister()
    → session added to registry
    → RunAutostart(projectPath)
      → load .agnt.kdl
      → port conflict pre-flight
      → TopologicalSort(scripts) → layers
      → for each layer: startAutostartScripts()
        → processReadiness.MarkStarting(pid)
        → waitForDependencies() → processReadiness.Wait(depPid)
        → autostartScript() → StartScript()

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.