Skip to content
Back to skills

Claude Code

ASecurity

Activate ONLY when the user explicitly mentions Claude Code, Claude CLI, claude command, Anthropic Claude Code, installing Claude Code, using Claude Code, or APX runtime claude-code. Do not activate for generic Claude model discussion.

  • 12 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added September 27, 2026
ai-agentsrustbash

Works with

  • claude code
  • cli

Security analysis

A96/100
  • mediumInstalls packages at runtime which could introduce malicious dependencies

Pro shows the line behind each finding and how to fix it

Scanned September 27, 2026

npx -y skills add agentprojectcontext/apx --skill claude-code --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Claude Code?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Claude Code
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/agentprojectcontext-claude-code/badge)](https://www.skillsdirectory.com/skills/agentprojectcontext-claude-code)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
scope: internal
name: claude-code
description: "Activate ONLY when the user explicitly mentions Claude Code, Claude CLI, claude command, Anthropic Claude Code, installing Claude Code, using Claude Code, or APX runtime claude-code. Do not activate for generic Claude model discussion."
homepage: https://docs.anthropic.com/en/docs/claude-code
---
# Claude Code CLI

Use this skill only for Claude Code CLI install, auth, usage, or APX runtime dispatch.

## Verify before acting

Check the local CLI first:

```bash
claude --version
claude --help
```

Do not invent flags. If a command is uncertain, inspect help for the exact subcommand before
running it.

## Install

Common install/update path:

```bash
npm install -g @anthropic-ai/claude-code
claude --version
```

Claude Code also exposes:

```bash
claude install
claude update
claude auth
```

Use `claude --help` to confirm current syntax.

## Non-interactive use

Prefer headless print mode:

```bash
claude -p "task" --output-format json
claude -p "task" --append-system-prompt "system instructions" --output-format json
```

For high-trust automation in an already sandboxed environment:

```bash
claude -p "task" --permission-mode bypassPermissions --output-format json
```

## List and resume sessions

Claude Code has no `--list`; `--resume` is always an interactive picker. To list
sessions non-interactively, use APX:

```bash
apx sessions list --engine claude --project <name>   # registered APX project
apx sessions list --engine claude --dir <path>       # any directory
```

This prints each session's id and title. To resume one (run from the project directory):

```bash
claude --continue                       # most recent session
claude -p --resume <session-id> "..."   # specific session, always with -p (print mode)
```

## Not here: agent-to-agent (a2a) hand-off

Reaching an APX agent and being answered back (the a2a relay: `apx send … --deliver`,
`engine=… session=…` identity, `apx session resume … --msg`) is a **harness-side**
concern. It lives in the `apx` skill that every coding CLI loads, and on the agent
side in **apx-runtime**. This driver covers only how APX operates the Claude
engine — keep a2a out of it.

## APX runtime

Run a project agent through Claude Code:

```bash
apx run <agent> --runtime claude-code "task"
```

If the task needs Telegram, tell Claude Code the exact APX command:

```bash
apx telegram send "message"
```

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…