一键导入
claw
Install the claw CLI tool — run NanoClaw agent containers from the command line without opening a chat app.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Install the claw CLI tool — run NanoClaw agent containers from the command line without opening a chat app.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Browse the web for any task — research topics, read articles, interact with web apps, fill forms, take screenshots, extract data, and test web pages. Use whenever a browser would be useful, not just when the user explicitly asks.
How to make authenticated API calls to external services. Covers api.sh, cred-exec, auth patterns, available credentials, cloning repos, SSH forwarding, and troubleshooting 401s. Read this before calling any external API or using a CLI tool that needs auth.
Rich content blocks for the ClawDad web UI. Use when responding to web channels (folder starts with "web_") to emit structured blocks — code, cards, tables, stats, progress bars, diffs, alerts, action buttons, files, forms — alongside standard markdown.
Create and manage agent groups and multi-agent teams. Use when the user wants to create a new group, create a team, list groups, delete a group, add/remove agents from a team, or troubleshoot group configuration.
File a forward-looking GitHub issue (enhancement, design debt, follow-up) against dfederspiel/clawdad with the established structured-proposal template. Use when work surfaces a follow-up that isn't a confirmed bug — proposed redesigns, deferred cleanup, dependent issues, capability gaps. NOT for filing bugs (use /report-bug) or for filing without context. Triggers on "file followup", "file a followup", "/file-followup", or "file as an issue" when the topic is forward-looking rather than a defect.
Audit the context layers for a group or agent. Reports which CLAUDE.md files are in play, lists loaded skills with line counts, and flags any skill over the ~150-line compaction threshold where content may be silently truncated. Use when debugging agent drift, before shipping a new skill, or when an agent seems to ignore rules you thought were loaded. Triggers on "context audit", "audit context", "check compaction", "why is the agent ignoring X", or "/context-audit".
| name | claw |
| description | Install the claw CLI tool — run NanoClaw agent containers from the command line without opening a chat app. |
claw is a Python CLI that sends prompts directly to a NanoClaw agent container from the terminal. It reads registered groups from the NanoClaw database, picks up secrets from .env, and pipes a JSON payload into a container run — no chat app required.
-g needed for most use)-s <session-id>--pipe) for scripting and piping--list-groupscontainer or docker runtime (or override with --runtime)-v) shows the command, redacted payload, and exit codenanoclaw-agent:latest)container (Apple Container, macOS 15+) or docker available in PATHRun this skill from within the NanoClaw directory. The script auto-detects its location, so the symlink always points to the right place.
mkdir -p scripts
cp "${CLAUDE_SKILL_DIR}/scripts/claw" scripts/claw
chmod +x scripts/claw
mkdir -p ~/bin
ln -sf "$(pwd)/scripts/claw" ~/bin/claw
Make sure ~/bin is in PATH. Add this to ~/.zshrc or ~/.bashrc if needed:
export PATH="$HOME/bin:$PATH"
Then reload the shell:
source ~/.zshrc # or ~/.bashrc
claw --list-groups
You should see registered groups. If NanoClaw isn't running or the database doesn't exist yet, the list will be empty — that's fine.
# Send a prompt to the main group
claw "What's on my calendar today?"
# Send to a specific group by name (fuzzy match)
claw -g "family" "Remind everyone about dinner at 7"
# Send to a group by exact JID
claw -j "120363336345536173@g.us" "Hello"
# Resume a previous session
claw -s abc123 "Continue where we left off"
# Read prompt from stdin
echo "Summarize this" | claw --pipe -g dev
# Pipe a file
cat report.txt | claw --pipe "Summarize this report"
# List all registered groups
claw --list-groups
# Force a specific runtime
claw --runtime docker "Hello"
# Use a custom image tag (e.g. after rebuilding with a new tag)
claw --image nanoclaw-agent:dev "Hello"
# Verbose mode (debug info, secrets redacted)
claw -v "Hello"
# Custom timeout for long-running tasks
claw --timeout 600 "Run the full analysis"
Install Docker Desktop or Apple Container (macOS 15+), or pass --runtime explicitly.
The script auto-detects your NanoClaw directory and reads .env from it. Check that the file exists and contains at least one of: CLAUDE_CODE_OAUTH_TOKEN, ANTHROPIC_API_KEY, ANTHROPIC_AUTH_TOKEN.
The default timeout is 300 seconds. For longer tasks, pass --timeout 600 (or higher). If the container consistently hangs, check that your nanoclaw-agent:latest image is up to date by running ./container/build.sh.
Run claw --list-groups to see what's registered. Group lookup does a fuzzy partial match on name and folder — if your query matches multiple groups, you'll get an error listing the ambiguous matches.
Containers run with --rm so they are automatically removed. If the agent crashes before emitting the output sentinel, claw falls back to printing raw stdout. Use -v to see what the container produced. Rebuild the image with ./container/build.sh if crashes are consistent.
If claw can't find your database or .env, set the NANOCLAW_DIR environment variable:
export NANOCLAW_DIR=/path/to/your/nanoclaw