Claude Skill MartBrowse skillsQuick linesLearn by videoTerminal guideWhat is a Skill?
 
← Back to list

Collaborating with Grok (Grok CLI bridge)

Delegate prototyping, debugging, code review and live web/X research to xAI's Grok CLI from Claude Code, with structured JSON results and multi-turn sessions.

Dev & CodingAdvanced★ 130⑂ 17AI score 9/10Last updated: Aug 17, 2026

What it does

  • Drives grok -p headlessly through scripts/grok_bridge.py, streaming progress to stderr and returning structured JSON (success, complete, model, SESSION_ID, tool_counts).
  • Keeps multi-turn continuity via --SESSION_ID (resume), --session-id (new named session) and --continue.
  • Defines explicit safety postures: read-only review (--tools "read_file,grep,list_dir"), web/X research (deny shell + edit tools), and implementation (isolated worktree + --always-approve, only with user consent).
  • Enables live web and X search through the coding model's backend web_search (default grok-4.5 as of CLI 0.2.93).

Who it's for

  • Developers who want a cross-model second opinion on designs, patches or bug hypotheses.
  • Teams offloading large repo analysis to a parallel agent with a read-only allowlist.
  • Anyone needing fresh web/X research inline with coding work.

Examples

  1. Read-only review: --cd "." --tools "read_file,grep,list_dir" --PROMPT "Review login() in src/auth.py and propose the smallest safe fix as a unified diff".
  2. Live research: --model grok-4.5 --disallowed-tools "run_terminal_cmd,search_replace" --PROMPT "Search web and X for <question>, cite source URLs".
  3. Two-turn flow: analyze a bug in turn 1, then reuse the returned SESSION_ID to request the patch in turn 2.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/appautomaton/agent-designer/HEAD/skills/collaborating-with-grok/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the skills/collaborating-with-grok folder from the GitHub repo appautomaton/agent-designer into my ~/.claude/skills/collaborating-with-grok/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/appautomaton/agent-designer.git /tmp/agent-designer && mkdir -p ~/.claude/skills && cp -r /tmp/agent-designer/skills/collaborating-with-grok ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Install and authenticate Grok CLI first: run grok login or export XAI_API_KEY.
  2. Clone the repo: git clone https://github.com/appautomaton/agent-designer.git
  3. Copy the skill: mkdir -p ~/.claude/skills && cp -r agent-designer/skills/collaborating-with-grok ~/.claude/skills/
  4. Pre-authorize the bridge in Claude Code by adding "Bash(python3 *collaborating-with-grok*bridge.py*)" to permissions.allow (and to sandbox.excludedCommands on sandboxed hosts so the child CLI keeps network access).
  5. Verify: python3 ~/.claude/skills/collaborating-with-grok/scripts/grok_bridge.py --help, then --list-models should return success: true.
  6. Restart Claude Code and ask something like "get a second opinion from Grok on this module" to trigger the skill.
View source on GitHub ↗