@mackding/codexclaw
BInstall and operate a Telegram bot that exposes Codex, MCP, GitHub subagents, repo switching, and minimal frontend dev-server control.
Install
agr install @mackding/codexclaw --target claudeWrites 16 files into .claude/skills/, pinned to git-403c0426.
- .claude/skills/codexclaw/.env.example
- .claude/skills/codexclaw/.gitignore
- .claude/skills/codexclaw/.prettierignore
- .claude/skills/codexclaw/.prettierrc.json
- .claude/skills/codexclaw/AGENTS.md
- .claude/skills/codexclaw/README.md
- .claude/skills/codexclaw/SECURITY.md
- .claude/skills/codexclaw/SKILL.md
- .claude/skills/codexclaw/ecosystem.config.cjs
- .claude/skills/codexclaw/ecosystem.config.ts
- .claude/skills/codexclaw/eslint.config.js
- .claude/skills/codexclaw/eslint.config.ts
- .claude/skills/codexclaw/llms.txt
- .claude/skills/codexclaw/package-lock.json
- .claude/skills/codexclaw/package.json
- .claude/skills/codexclaw/tsconfig.json
Document
name: CodexClaw description: Install and operate a Telegram bot that exposes Codex, MCP, GitHub subagents, repo switching, and minimal frontend dev-server control.
CodexClaw
What This Skill Does
- runs Codex through Telegram
- keeps coding sessions scoped to
chat + repo - exposes
/repo,/status,/skill,/gh,/mcp, and/dev - supports frontend repo debugging with
/dev start|stop|status|logs|url
Install
git clone https://github.com/MackDing/CodexClaw.git
cd CodexClaw
npm install
cp .env.example .env
Required Env
Set at least:
BOT_TOKEN=123456789:telegram-token
ALLOWED_USER_IDS=123456789
STATE_FILE=.codex-telegram-claws-state.json
WORKSPACE_ROOT=.
CODEX_WORKDIR=.
CODEX_BACKEND=sdk
Start
npm run start
Verify
npm run check
npm run lint
npm run format:check
npm test
npm run healthcheck
Telegram Quick Use
/status
/repo
/skill
/repo my-project
/dev status
/gh create repo my-new-repo
/gh confirm
Frontend Debugging
Use these commands inside the current repo selected by /repo:
/dev start/dev stop/dev status/dev logs/dev url/gh create repo ...,/gh push, and/gh commit "..."are explicit write paths/gh confirmexecutes the pending GitHub write action- plain-text
create repo,commit, andpushrequests are blocked and turned into guidance
Rules:
devscript is preferredstartscript is used as fallback- frontend runtime is shared per repo, not per chat
/devis not a general-purpose shell
Notes
superpowersis an internal workflow and shows up in/status, not as a toggleable/skill/shremains a restricted operator channel and is separate from/dev
Trustgrade B
- passBody integrity
Whether the stored document is plausibly the kind of file the artifact declares, rather than something fetched by mistake.
- passType matchnot applicable to this artifact type
Whether the artifact is really the kind of thing its metadata claims it is.
- passFreshness
How long since the source repository was last pushed to.
- passPrompt injection
Scans the artifact's own text for instructions aimed at your agent rather than at you.
- warnLicenseno SPDX license detected
Whether the source repository declares an SPDX license permissive enough to redistribute.
How the grade is calculated
Each check contributes 0 points when it passes, 1 when it warns, and 2 when it fails. The total maps to a letter:
- Aevery check passed
- Bone warning
- Ctwo warnings
- Dprompt injection or body integrity failed, or three warnings
- Fone of those failed, and something else is wrong
These are automated hygiene checks, not a security audit, and not a dependency or vulnerability scan. A grade of A means nothing was flagged — not that the artifact is safe.
Versions
git-403c04263f382026-07-31