Updated 2mo ago
by mfielding92
ClawedBack
🦀 A port of OpenClaw to Claude Code — personal AI assistant built with skills + minimal Python
23 GitHub stars (23)
Steady at 23 stars — no new stars in the 16 days since Sep 30.
- 30 skills (30 skills found in its SKILL.md files)
- 0 forks (0 forks)
- 2mo ago last update (Last updated 2mo ago)
Install
Skills CLI
npx skills add mfielding92/ClawedBack
Asks which skills and agents to install. Just one: npx skills add mfielding92/ClawedBack --skill code-review — add -g for every project.
Claude Code
git clone https://github.com/mfielding92/ClawedBackcp -r "ClawedBack/.claude/skills/code-review" ~/.claude/skills/
Copies the code-review skill into ~/.claude/skills/ for every project (use .claude/skills/ for one project). Repeat for the other skills listed below.
Plugin
/plugin marketplace add mfielding92/ClawedBack
Run inside Claude Code if the repository ships a plugin marketplace, then pick plugins from /plugin.
Skills can run code on your machine — read the SKILL.md first.
What it does
In the author's own words, from each skill's SKILL.md — showing the first 30 of 32 files.
-
code-review
Review code changes for quality, security, and correctness. Use when the user says "review this PR", "review these changes", "check my code", "look at what I changed", or after implementing a feature. Produces a severity-organized report.
.claude/skills/code-review/SKILL.md on GitHub (opens in a new tab) -
context-doctor
Diagnose and fix context window issues. Use when Claude seems confused, forgetful, slow, or the user says "context is full", "you forgot", "you're repeating yourself", "start fresh", or "why are you ignoring my instructions".
.claude/skills/context-doctor/SKILL.md on GitHub (opens in a new tab) -
git-workflow
Structured git workflow for commits, branches, and PRs. Use when the user says "commit", "create a PR", "new branch", "prepare for review", "ship this", or "push my changes". Safe, conventional git operations with checkpoints.
.claude/skills/git-workflow/SKILL.md on GitHub (opens in a new tab) -
grill-me
Deep interview to reach shared understanding before building. Use when starting a complex feature, when requirements are unclear, or when the user says "grill me", "interview me", "ask me questions", "let's figure this out", or "grill me lightly" for a quick version.
.claude/skills/grill-me/SKILL.md on GitHub (opens in a new tab) -
oc-approve
Approval gate for dangerous operations. Use internally before executing destructive commands, file deletions, git pushes, or external API calls from the web chat. Do NOT invoke directly.
.claude/skills/oc-approve/SKILL.md on GitHub (opens in a new tab) -
oc-automate
Create, list, and manage scheduled tasks and automations. Use when the user asks to schedule something, set a reminder, create a recurring task, or manage existing automations. Trigger phrases: 'remind me', 'every hour', 'schedule', 'at 9am', 'recurring', 'automate', 'cron'.
.claude/skills/oc-automate/SKILL.md on GitHub (opens in a new tab) -
oc-channel
Manage communication channel adapters (web chat, Telegram, Discord, etc.). Use when the user wants to add a new channel, list channels, or configure channel settings. Trigger phrases: 'add channel', 'connect telegram', 'connect discord', 'list channels'.
.claude/skills/oc-channel/SKILL.md on GitHub (opens in a new tab) -
oc-crosstalk
Communicate with another ClawedBack instance. Use when the user says 'crosstalk', 'talk to server', 'contact instance', 'send message to', or '/oc-crosstalk'. Enables ClawedBack-to-ClawedBack communication.
.claude/skills/oc-crosstalk/SKILL.md on GitHub (opens in a new tab) -
oc-files
Manage files and create temporary share links for the web chat. Use when the user says 'share file', 'send file', 'list files', 'download', or when you need to provide a file to the user via the chat. Also use when you generate a file (PDF, image, code, etc.) and want to let the user download it.
.claude/skills/oc-files/SKILL.md on GitHub (opens in a new tab) -
oc-hub
Browse, install, and import skills for clawed-back. Supports local skills AND importing OpenClaw/ClawHub skills (they share the AgentSkills format). Use when the user says 'install skill', 'list skills', 'what skills are available', 'skill store', 'marketplace', 'hub', 'clawhub', 'import skill', 'openclaw skill'.
.claude/skills/oc-hub/SKILL.md on GitHub (opens in a new tab) -
oc-local
Run tasks on a local or remote Ollama model instead of Claude. Use when the user says 'run locally', 'use ollama', 'local model', 'process offline', 'use local', '/oc-local', or wants to offload a task to a local LLM.
.claude/skills/oc-local/SKILL.md on GitHub (opens in a new tab) -
oc-poll
Check the message queue for new user messages and process them. This is the heartbeat of clawed-back. Typically triggered by CronCreate on a schedule.
.claude/skills/oc-poll/SKILL.md on GitHub (opens in a new tab) -
oc-report
Internal bug report generator. Produces PII-redacted, structured reports formatted as GitHub issues. Called by other skills when errors are detected — not directly by users.
.claude/skills/oc-report/SKILL.md on GitHub (opens in a new tab) -
oc-respond
Write a response to the user via the message queue. Use internally when processing a user message and you need to send the reply back through the web chat. Do NOT use for direct terminal conversations.
.claude/skills/oc-respond/SKILL.md on GitHub (opens in a new tab) -
oc-resume
Resume ClawedBack after a restart. Starts the web server, resumes polling, and rebuilds config if missing. Use when starting a new Claude session in an existing ClawedBack install, or when the user says 'resume', 'start', 'restart', 'begin polling', or 'oc-resume'.
.claude/skills/oc-resume/SKILL.md on GitHub (opens in a new tab) -
oc-router
Route and process incoming user messages from the web chat queue. Use internally when oc-poll detects new messages. Do NOT invoke directly.
.claude/skills/oc-router/SKILL.md on GitHub (opens in a new tab) -
oc-session
Load and save conversation session state. Use internally when processing messages to maintain context across the chat. Do NOT invoke directly.
.claude/skills/oc-session/SKILL.md on GitHub (opens in a new tab) -
oc-setup
First-run setup wizard for clawed-back. Use when the user says 'setup', 'start clawed-back', 'initialize', or when running for the first time. Offers three install modes: portable user, portable root, or system install.
.claude/skills/oc-setup/SKILL.md on GitHub (opens in a new tab) -
oc-ssl
Set up SSL/HTTPS certificates using Let's Encrypt certbot. Use when the user says 'setup ssl', 'get certificate', 'enable https', 'certbot', 'letsencrypt', or wants to secure their ClawedBack instance with HTTPS on a public domain.
.claude/skills/oc-ssl/SKILL.md on GitHub (opens in a new tab) -
oc-token
View, regenerate, or set the auth token for the web chat. Use when the user says 'change token', 'reset token', 'show token', 'set password', 'new token', or 'auth token'.
.claude/skills/oc-token/SKILL.md on GitHub (opens in a new tab) -
oc-tools
Execute tools and commands on behalf of the user from the web chat. Use internally when the router identifies a tool request. Do NOT invoke directly.
.claude/skills/oc-tools/SKILL.md on GitHub (opens in a new tab) -
oc-update
Pull the latest ClawedBack updates from GitHub and apply them to the running installation, preserving local config, data, auth tokens, and user-installed skills. Use when the user says 'update', 'upgrade', 'pull latest', 'check for updates', or 'oc-update'.
.claude/skills/oc-update/SKILL.md on GitHub (opens in a new tab) -
oc-voice
Handle voice message processing. Use internally when the router encounters a voice-type message. Do NOT invoke directly.
.claude/skills/oc-voice/SKILL.md on GitHub (opens in a new tab) -
oc-webhook
Handle incoming webhook payloads from external services. Use internally when the router encounters a webhook-type message. Supports GitHub, GitLab, custom webhooks. Do NOT invoke directly.
.claude/skills/oc-webhook/SKILL.md on GitHub (opens in a new tab) -
plan-and-spec
Create a structured implementation plan with spec-driven modeling. Use when starting a new feature, complex refactor, or when the user says "plan this", "spec this out", "design this", or "how should we build this". Forces multiple design iterations before coding.
.claude/skills/plan-and-spec/SKILL.md on GitHub (opens in a new tab) -
progress-tracker
Create and update progress/handoff documents for long-running work. Use when nearing context limits, switching tasks, ending a session, or when the user says "save progress", "handoff", "wrap up", or "I need to pick this up later".
.claude/skills/progress-tracker/SKILL.md on GitHub (opens in a new tab) -
refactor
Safely refactor code with characterization tests and incremental changes. Use when the user says "refactor", "clean up", "extract", "rename", "restructure", or needs to modernize legacy code.
.claude/skills/refactor/SKILL.md on GitHub (opens in a new tab) -
research
Structured web research with source evaluation and synthesis. Use when the user says "research", "look up", "find out", "what's the best way to", "how do people", "find documentation for", or needs information from the web to make a decision.
.claude/skills/research/SKILL.md on GitHub (opens in a new tab) -
robust-edit
Robust, line-based file editing. Replace a range of lines in a file with new content by line number instead of exact string matching. Use when the user says 'robust-edit', '/robust-edit', 'replace lines', 'edit lines N to M', or when character-level matching with the Edit tool keeps failing on multi-line changes or uncertain whitespace.
.claude/skills/robust-edit/SKILL.md on GitHub (opens in a new tab) -
start
Start ClawedBack — ensures the web server is running, creates the polling cron job, and runs one poll cycle. Use when starting or restarting Claude Code in a ClawedBack project, or when the user says 'start', '/start', 'begin', or 'get going'.
.claude/skills/start/SKILL.md on GitHub (opens in a new tab)
No skill matches that filter.
Review before you install
A skill is a set of instructions and scripts your agent will follow, and it can run code on your machine. Read the SKILL.md and any scripts first — this listing is not a security review.