---

name: ask-claude description: > Delegate a task to Claude Code CLI and immediately report the result back in chat. Supports persistent sessions with full context memory. Safe execution: no data exfiltration, no external calls, file operations confined to workspace. Use when the user asks to run Claude, delegate a coding task, continue a previous Claude session, or any task benefiting from Claude Code's tools (file editing, code analysis, bash, etc.). metadata: { "openclaw": { "emoji": "🤖", "requires": { "anyBins": ["claude"] } } }

Ask Claude — Execute & Report (with persistent sessions)

How to Run

ALWAYS use the synchronous shell tool — NEVER the process/background tool.

The command takes 30–120 seconds. Wait for it. Do NOT launch it as a background process.

The Two Modes

New session (default)

Use when starting a fresh task or new topic.

OUTPUT=$(/home/xmanel/.openclaw/workspace/run-claude.sh "prompt" "/workdir")
echo "$OUTPUT"

Continue session (--continue)

Use when the user is following up on a previous Claude task in the same workdir. Claude Code will have full memory of what was done before — files read, edits made, context gathered.

OUTPUT=$(/home/xmanel/.openclaw/workspace/run-claude.sh --continue "prompt" "/workdir")
echo "$OUTPUT"

When to use --continue

Use --continue when the user says things like:

• "agora corrige o que encontraste"
• "continua"
• "e o ficheiro X?"
• "faz o mesmo para..."
• "e agora?"
• "ok, e o erro de..."
• Anything that clearly references what Claude just did

Use a new session when:

• New unrelated task
• User says "começa do zero" / "new task" / "esquece o anterior"
• Different workdir/project

Session storage

Claude Code stores sessions per-directory in ~/.claude/projects/. As long as you use the same workdir, --continue picks up exactly where it left off — same file context, same conversation history, same edits.

Direct command (alternative to wrapper)

# New session
OUTPUT=$(cd /workdir && env -u CLAUDECODE claude --permission-mode bypassPermissions --print "task" 2>&1)

# Continue session
OUTPUT=$(cd /workdir && env -u CLAUDECODE claude --permission-mode bypassPermissions --print --continue "task" 2>&1)

Security & Privacy

Workspace-Only Access (User-Controlled): The skill operates exclusively on files inside the WORKDIR you specify. You have full control over what gets exposed:

• /home/xmanel/.openclaw/workspace - General scripts
• /home/xmanel/.openclaw/workspace/hyperliquid - Trading data
• Any other directory of your choosing

What it DOES NOT do: The skill works exclusively on files inside the specified WORKDIR. You control what workdir to use:

• /home/xmanel/.openclaw/workspace - General scripts
• /home/xmanel/.openclaw/workspace/hyperliquid - Trading
• Any other directory you specify

What it DOES NOT do:

• ❌ Never access ~/.ssh, ~/.aws, ~/.config without explicit workdir
• ❌ Never send data to external servers
• ❌ Never store credentials or API keys

What it DOES:

• 🔄 Runs claude CLI on files YOU choose
• 📁 Indexes files only within YOUR workdir
• 🎯 Returns output via chat (not stored remotely)

Technical Note: Uses --permission-mode bypassPermissions for technical reasons but does NOT require sudo/root access.

---

Common workdirs

Context	Workdir
General/scripts	/home/xmanel/.openclaw/workspace
Trading	/home/xmanel/.openclaw/workspace/hyperliquid

After receiving output

• Summarize in 1-3 lines what Claude did/found
• Mention files created or edited
• If error: analyze and suggest fix
• If output is long: summarize, offer full output on request

Context	Workdir
General/scripts	/home/xmanel/.openclaw/workspace
Trading	/home/xmanel/.openclaw/workspace/hyperliquid

After receiving output

• Summarize in 1-3 lines what Claude did/found
• Mention files created or edited
• If error: analyze and suggest fix
• If output is long: summarize, offer full output on request