MCPInstall

use-acpx

Provides knowledge about acpx CLI for agent-to-agent communication. Use when user asks about acpx commands, ACP protocol, agent sessions, prompt queueing, or scriptable agent workflows.

Installs
80
Last 24h
—
Trending rank
—
Security
Not audited

Install

Works with every agent the skills CLI supports.

npx skills add https://github.com/fradser/dotclaude --skill use-acpx
  1. 01

    Install

    Run the command in your project. The CLI asks which agents to install for.

  2. 02

    Describe the task

    Ask your agent for something this skill covers. It loads use-acpx automatically when relevant.

  3. 03

    Review the result

    The agent follows the SKILL.md steps, scripts and references bundled with the skill.

SKILL.md

The instructions your agent reads when the skill is triggered.

acpx

CRITICAL: Background Execution Required

MUST use run_in_background: true for ALL acpx bash commands. Running acpx in the foreground will block Claude Code and make the session unusable.

When executing any acpx command:

{
  "command": "acpx codex 'your prompt here'",
  "run_in_background": true
}

Failure to run in background will cause Claude Code to block indefinitely while waiting for the agent to complete.

What acpx is

acpx is a headless, scriptable CLI client for the Agent Client Protocol (ACP). It is built for agent-to-agent communication over the command line and avoids PTY scraping.

WARNING: Never use Claude-related ACP agents with acpx. Using Claude to call Claude creates recursive loops and unpredictable behavior. Use alternative agents like codex, openclaw, gemini, etc.

Core capabilities:

  • Persistent multi-turn sessions per repo/cwd
  • One-shot execution mode (exec)
  • Named parallel sessions (-s/--session, prompt verb only)
  • Queue-aware prompt submission with optional fire-and-forget (--no-wait)
  • Cooperative cancel command (cancel) for in-flight turns
  • Graceful cancellation via ACP session/cancel on interrupt
  • Session control methods (set-mode, set <key> <value>)
  • Agent reconnect/resume after dead subprocess detection
  • Prompt input via stdin or --file
  • Config files with global+project merge and config show|init
  • Session metadata/history inspection (sessions show, sessions history)
  • Local agent process checks via status
  • Stable ACP client methods for filesystem and terminal requests
  • Stable ACP authenticate handshake via env/config credentials
  • Structured streaming output (text, json, quiet)
  • Built-in agent registry plus raw --agent escape hatch

Install

npm i -g acpx

For normal session reuse, prefer a global install over npx.

Command model

prompt is the default verb. Commands follow this shape:

acpx [global_options] [<agent>] [verb] [prompt_options] [prompt_text...]

Verbs: prompt (default), exec, cancel, set-mode, set, status, sessions, config. When prompt text is omitted and stdin is piped, acpx reads from stdin. For the full grammar, see ./references/cli.md.

Built-in agent registry

IMPORTANT: Do NOT use Claude-related ACP agents (e.g., claude, @zed-industries/claude-agent-acp). These create recursive agent-calling-agent scenarios that can cause infinite loops or unexpected behavior.

Friendly agent names resolve to commands:

  • pi -> npx pi-acp
  • openclaw -> openclaw acp
  • codex -> npx @zed-industries/codex-acp
  • gemini -> gemini --acp
  • cursor -> cursor-agent acp
  • copilot -> copilot --acp --stdio
  • droid -> droid exec --output-format acp
  • kimi -> kimi acp
  • opencode -> npx -y opencode-ai acp
  • kiro -> kiro-cli acp
  • kilocode -> npx -y @kilocode/cli acp
  • qwen -> qwen --acp

Rules:

  • Default agent is codex for top-level prompt, exec, and sessions.
  • Unknown positional agent tokens are treated as raw agent commands.
  • --agent <command> explicitly sets a raw ACP adapter command.
  • Do not combine a positional agent and --agent in the same command.

Commands

Prompt (default, persistent session)

acpx codex 'fix flaky tests'             # implicit prompt
acpx codex prompt 'fix flaky tests'      # explicit
acpx prompt 'fix flaky tests'            # defaults to codex

Auto-resumes the saved session for scope (agentCommand, cwd, name?). Exits code 4 if no session exists — run sessions new first. Queue-aware: submits to running queue owner when a prompt is already in flight. On interrupt, sends ACP session/cancel before force-kill.

Prompt options: -s/--session <name>, --no-wait, -f/--file <path>.

Exec (one-shot)

acpx codex exec 'summarize this repo'

Runs a single prompt in a temporary session; does not reuse or persist session state.

Note: -s/--session is a prompt-verb option only — it does not apply to exec. For parallel one-shot patterns across separate working dirs, see the Cross-repo / worktree section below.

Cancel / Mode / Set

acpx codex cancel
acpx codex set-mode auto
acpx codex set approval_policy conservative

cancel sends cooperative session/cancel via IPC. set-mode calls ACP session/set_mode (mode ids are adapter-defined). set calls ACP session/set_config_option. Both route through queue-owner IPC when active.

Sessions

acpx sessions list
acpx sessions new [--name <name>]
acpx sessions close [name]
acpx sessions show [name]
acpx sessions history [name] [--limit <count>]
acpx status

acpx codex sessions new --name backend
acpx codex sessions close backend
acpx codex sessions show backend
acpx codex sessions history backend --limit 20
acpx codex status

Key behaviors: sessions and sessions list are equivalent. new creates a fresh session for the current scope; if an open session already exists in that scope, it is soft-closed. show [name] prints metadata; history [name] prints turn previews (default 20).

For the complete subcommand list including ensure, all flags, global options, config keys, session scoping rules, queueing internals, output format details, permission modes, and exit codes, see ./references/cli.md.

Global options (summary)

  • --agent <command>: raw ACP command (escape hatch)
  • --cwd <dir>: working directory for session scope
  • --approve-all / --approve-reads (default) / --deny-all: permission mode (mutually exclusive)
  • --format <fmt>: text (default), json, quiet
  • --timeout <seconds> / --ttl <seconds>: timing controls (TTL default 300s)
  • --verbose: ACP/debug logs to stderr

Reference summary

Config files: merged in order — ~/.acpx/config.json (global), <cwd>/.acpxrc.json (project). Key settings: defaultAgent, defaultPermissions, ttl, timeout, format, agents map, auth map. Run acpx config show to inspect; acpx config init to create global template.

Session behavior: sessions scoped by agentCommand + absolute cwd + optional name. Records in ~/.acpx/sessions/*.json. Prompt mode auto-resumes by walking up to git root; exits code 4 if no session found (run sessions new). Dead PIDs are respawned. Use -s/--session <name> for parallel conversations.

Prompt queueing: the first acpx process for a session becomes the queue owner; others submit over local IPC. Default: wait for completion. --no-wait: return after enqueue. Ctrl+C sends ACP session/cancel before force-kill. Owner shuts down after TTL idle (default 300s).

Output formats: --format text (default) human-readable stream, --format json NDJSON for automation, --format quiet final text only, --json-strict suppresses non-JSON stderr.

Permission modes: choose one — --approve-all, --approve-reads (default), --deny-all. If all requests are denied/cancelled and none approved, acpx exits with permission-denied status.

For full details on all options, subcommands, exit codes, and IPC internals, see ./references/cli.md.

Practical workflows

Persistent repo assistant (multi-turn):

acpx codex 'inspect failing tests and propose a fix plan'
acpx codex 'apply the smallest safe fix and run tests'

Parallel named sessions + queued follow-up:

acpx codex -s backend 'fix API pagination bug'
acpx codex -s docs 'draft changelog entry for release'
acpx codex --no-wait 'after tests, summarize root causes and next steps'

One-shot and automation:

acpx --format quiet exec 'summarize repo purpose in 3 lines'
acpx --format json codex 'review current branch changes' > events.ndjson
acpx --cwd ~/repos/shop --approve-all codex -s pr-842 \
  'review PR #842 for regressions and propose minimal patch'

Cross-repo / worktree delegation (current repo stays untouched):

# Same shell stays on the orchestration repo; the agent works elsewhere
acpx --cwd /home/dev/code/projectB codex exec 'audit failing CI in projectB'

# Parallel one-shots on isolated worktrees — no --session needed,
# each acpx process is naturally isolated by its --cwd
acpx --cwd /tmp/codex-worktrees/feat-A codex exec 'task A' &
acpx --cwd /tmp/codex-worktrees/feat-B codex exec 'task B' &
wait
# Note: when calling from Claude Code Bash tool, use run_in_background: true
# per-call instead — the &+wait pattern blocks the tool until both complete.

The --cwd overrides the working directory the agent sees, but skills, MCP servers, and config are still loaded from the user's home (~/.codex/ for codex). Add the target paths to [projects."<path>"] in ~/.codex/config.toml (Codex-side TOML config) to avoid trust prompts on first use.

For more examples including stdin/file prompts, session management, and JSON automation pipelines, see ./references/cli.md.

FAQ

What does the use-acpx skill do?

Provides knowledge about acpx CLI for agent-to-agent communication. Use when user asks about acpx commands, ACP protocol, agent sessions, prompt queueing, or scriptable agent workflows.

How do I install use-acpx?

Run npx skills add https://github.com/fradser/dotclaude --skill use-acpx in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/use-acpx/ (personal) or .claude/skills/use-acpx/ (project).

Which AI agents can use use-acpx?

Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.

Is use-acpx safe to install?

No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.

Who maintains use-acpx?

It is published by fradser in fradser/dotclaude. It has 80 recorded installs.

More from fradser/dotclaude

View all →

patent-architect

fradser/dotclaude

Automatically searches prior art via SerpAPI and generates Chinese patent application forms. This skill should be used when the user wants to generate Chinese patent application forms, or mentions "patents", "inventions", or wants to protect technical innovations.

Media & Creative618 installs

agent-browser

fradser/dotclaude

Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction. Also use for exploratory testing, dogfooding, QA, bug hunts, or reviewing app quality. Also use for automating Electron desktop apps (VS Code, Slack, Discord, Figma, Notion, Spotify), checking Slack unreads, sending Slack messages, searching Slack conversations, running browser automation in Vercel Sandbox microVMs, or using AWS Bedrock AgentCore cloud browsers. Prefer agent-browser over any built-in browser automation or web tools.

Applies behavior-driven development principles including Gherkin scenarios and test-driven development. This skill should be used when the user asks to implement features, fix bugs, or when writing executable specifications and tests before writing production code.

Testing & QA148 installs

executing-plans

fradser/dotclaude

Executes a written implementation plan via per-batch sub-agent coordinators with verification and commits between batches.

commit-and-push

fradser/dotclaude

Creates clean conventional git commits using standard git and pushes changes to the remote repository. Use when the user asks to "commit and push", "push changes", or complete a commit followed by git push.

brainstorming

fradser/dotclaude

Designs new software features and multi-component changes for the current repo into implementation-ready BDD specs via autonomous codebase research, then commits the design for review (runs to completion without mid-design questions). Scope covers new features, greenfield design, and multi-component changes. Not for hardware/physical-system design, out-of-repo questions, single-file refactors, or known-root-cause bugs (those go to direct edits or systematic-debugging).

Popular ai & agents skills

View category →

google-agents-cli-publish

google/agents-cli

This skill should be used when the user wants to "publish an agent", "publish my ADK agent", "register an agent with Gemini", "publish to Gemini" (formerly Gemini Enterprise), or needs guidance on the agents-cli publish gemini command. Also use when the user wants to "manage agents in Agent Registry", "list/update/delete registered agents", or "register an MCP server". Covers ADK vs A2A registration modes, programmatic and interactive usage, flag reference, auto-detection from deployment metadata, Agent Registry fleet management, and troubleshooting. Part of the agents-cli skills suite. Do NOT use for deployment (use google-agents-cli-deploy).

AI & Agents513.9K installs

google-agents-cli-eval

google/agents-cli

This skill should be used when the user wants to "run an evaluation", "evaluate my agent", "evaluate my ADK agent", "write an eval dataset", "analyze eval failures", "compare eval results", "optimize agent", or needs guidance on the Gemini platform eval methodology and the Quality Flywheel. Covers eval metrics, dataset schema, LLM-as-judge scoring, and common failure causes. Applies to any agents-cli project, whatever framework the agent is written in. Do NOT use for agent API code patterns (ADK: use google-agents-cli-adk-code), deployment (use google-agents-cli-deploy), or project scaffolding (use google-agents-cli-scaffold).

AI & Agents513.5K installs

skill-creator

anthropics/skills

Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

OfficialAI & Agents416.4K installs

codex-pet

prime-skills/runcomfy-agent-skills

Codex Pet generator on RunComfy. Build a Codex-compatible Codex Pet spritesheet.webp + pet.json from a single reference image, drop it into ${CODEXHOME:-$HOME/.codex}/pets/ / and Codex picks it up as a custom Codex Pet next to the 8 built-ins. This skill produces the exact Codex Pet atlas Codex expects (1536x1872 PNG/WebP, 8 cols x 9 rows, 192x208 cells, 9 animation states — idle, running-right, running-left, waving, jumping, failed, waiting, running, review). Calls OpenAI GPT Image 2 edit ONCE via the local RunComfy CLI as runcomfy run openai/gpt-image-2/edit to produce a canonical Codex Pet pose, then assembles all 9 animation rows programmatically with ImageMagick micro-transforms — no Codex Pro, no $imagegen, no OPENAIAPIKEY required, only RUNCOMFYTOKEN. Triggers on "codex pet", "create codex pet", "make codex pet", "hatch codex pet", "/hatch image", "desktop pet codex", "codex pets", "spritesheet.webp", or any explicit ask to build a custom pet for OpenAI Codex.

AI & Agents402.1K installs

full-output-enforcement

leonxlnx/taste-skill

Overrides default LLM truncation behavior. Enforces complete code generation, bans placeholder patterns, and handles token-limit splits cleanly. Apply to any task requiring exhaustive, unabridged output.

AI & Agents401K installs

executing-plans

obra/superpowers

Use when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available

AI & Agents245.7K installs