jackson2w/claude-code-skills / archive/openclaw-deployment
openclaw-deployment
FlaggedThis skill should be used when deploying or debugging a self-hosted OpenClaw (formerly Clawdbot/Moltbot) agent gateway — a Node.js personal-assistant service bridging Telegram/WhatsApp/Discord to an LLM with tool/skill access. Covers install, secure baseline config (loopback bind, token auth, exec approval gating), the two non-obvious config traps that silently break things (model selection vs. API key, exec security vs. ask), Telegram channel lockdown, configuring the web_search tool/Brave provider, voice-note transcription (tools.media.audio/Groq), memory indexing and its embedding-provider auth (openclaw memory index/status), evaluating ClawHub skills/plugins by real usage data, and building trustworthy custom skills instead of pulling from the unvetted ClawHub marketplace. Trigger phrases include "openclaw config", "openclaw gateway", "tools.exec.security", "openclaw model not found", "ProviderAuthError No API key found for provider openai", "exec denied security=deny", "openclaw dashboard", "openclaw pairing", "ClawHub skill", "openclaw skills install local", "openclaw web search", "tools.web.search.provider", "web_search provider is not available", "openclaw plugins install clawhub", "plugins.allow is empty", "tools.media.audio", "voice memo transcription", "openclaw skills search --json", "ClawHub popular skills", "EXTERNALLY-MANAGED pip openclaw", "AgentMail openclaw", "clawdbot paths in skill", "openclaw memory index", "memorySearch.provider", "memory index failed no api key", "credit_balance_exhausted", "insufficient_quota embeddings", "memory index --force", "scoped sudo for openclaw agent", "dispatch script security boundary", "openclaw agent --message-file", "openclaw agent --session-key timeout", "brief openclaw on new capabilities", "openclaw treats primer as prompt injection", "openclaw session jsonl toolCall", "arm olu with new tools", "expand openclaw agent privilege incrementally", "skill_workshop", "openclaw agent rejects everything as injected",
Security report
FlaggedStatic analysis of one exact version. It reduces risk and shows evidence; it does not certify that a skill is safe.
- Commit
df337688e763edd4364e24e577c1f1321bef6264- Content hash
c53fbc558d1bbd67084a6f97e9dd52c4e5f02d788d4da8f734bf4338e86c06f6- Scanned
- 2026-11-17
- Scanner version
- skillxone-scan-1.0.0
- Policy version
- 0.1-draft
- Decision
- 2 open critical or high findings.
- Verification expires
- Not applicable
Findings (2 critical, 15 low, 4 info)
The command gives a remote host interactive control of this machine (a reverse shell).
sudo -u openclaw timeout 5 bash -c 'echo > /dev/tcp/<some-other-internal-host>/22' \Suggested fix: Remove it.
The skill refers to system password files, a location that holds private data or credentials. The same file also sends data over the network, so severity was raised one level.
# /etc/sudoers.d/openclaw-remoteSuggested fix: Do not read credential stores or private files. If the task needs a credential, ask the user to provide it through the environment.
- low
INT-MISSING-LICENSENeither the skill nor its repository states a license, so the terms of use are unclear.
Suggested fix: Add a LICENSE file to the repository or a `license` field to SKILL.md.
A command runs with administrator privileges.
`/proc/<pid>/mountinfo`), not a `sudo -u openclaw` shortcut — see "`NoNewPrivileges=yes`Suggested fix: Avoid sudo where possible. If it is required, explain why so the user can decide.
A command runs with administrator privileges.
sudo systemctl restart openclaw.service # plugin install requires a gateway restart to loadSuggested fix: Avoid sudo where possible. If it is required, explain why so the user can decide.
A command runs with administrator privileges.
production, source the env file and fix `HOME` explicitly (`sudo -E` alone doesn't fix `HOME` whenSuggested fix: Avoid sudo where possible. If it is required, explain why so the user can decide.
A command runs with administrator privileges.
sudo bash -c 'set -a; source /root/.config/openclaw-*.env; set +a; \Suggested fix: Avoid sudo where possible. If it is required, explain why so the user can decide.
A command runs with administrator privileges.
sudo -u openclaw HOME=/home/openclaw openclaw config validate'Suggested fix: Avoid sudo where possible. If it is required, explain why so the user can decide.
A command runs with administrator privileges.
(`sudo cat /proc/$(systemctl show openclaw.service -p MainPID --value)/environ | tr '\0' '\n'`Suggested fix: Avoid sudo where possible. If it is required, explain why so the user can decide.
A command runs with administrator privileges.
sudo awk '{ if (match($0, /^[A-Za-z_]+=/)) print substr($0,RSTART,RLENGTH) "[...]"; \Suggested fix: Avoid sudo where possible. If it is required, explain why so the user can decide.
A command runs with administrator privileges.
`sudo systemctl daemon-reload && sudo systemctl restart openclaw.service` to apply. Verify withSuggested fix: Avoid sudo where possible. If it is required, explain why so the user can decide.
A command runs with administrator privileges.
Instead grant `NOPASSWD` sudo to exactly one root-owned script path per capability tier, and putSuggested fix: Avoid sudo where possible. If it is required, explain why so the user can decide.
A command runs with administrator privileges.
same script.** Confirmed 2026-08-18 auditing Olu's exec failures: `sudo bash -xSuggested fix: Avoid sudo where possible. If it is required, explain why so the user can decide.
A command runs with administrator privileges.
### This sudo pattern works for `openclaw-remote` over SSH, and CANNOT work for the `openclaw` service accountSuggested fix: Avoid sudo where possible. If it is required, explain why so the user can decide.
A command runs with administrator privileges.
escalation before sudo ever reads sudoers, so any grant to it is decorative. The error is a bareSuggested fix: Avoid sudo where possible. If it is required, explain why so the user can decide.
A command runs with administrator privileges.
process, which is usually what the sudo grant was reaching for anyway. Two companions:Suggested fix: Avoid sudo where possible. If it is required, explain why so the user can decide.
A command runs with administrator privileges.
looks correct (`sudo -l -U openclaw` shows it) but **is dead on arrival** — the kernel-levelSuggested fix: Avoid sudo where possible. If it is required, explain why so the user can decide.
A command runs with administrator privileges. Severity lowered from low because the surrounding text warns against it.
description: This skill should be used when deploying or debugging a self-hosted OpenClaw (formerly Clawdbot/Moltbot) agent gateway — a Node.js personal-assistant service bridging Telegram/WhatsApp/Discord to an LLM with tool/skill access.…Suggested fix: Avoid sudo where possible. If it is required, explain why so the user can decide.
The skill sends data to a chat webhook, which accepts arbitrary data. Severity lowered from low because the surrounding text warns against it.
curl "https://api.telegram.org/bot$TELEGRAM_BOT_TOKEN/deleteMessage" \Suggested fix: Make sure only the intended message is sent and that the user knows about it.
A command runs with administrator privileges. Severity lowered from low because the surrounding text warns against it.
process's own writes; a separate root-owned script invoked via sudo was never inside thatSuggested fix: Avoid sudo where possible. If it is required, explain why so the user can decide.
A command runs with administrator privileges. Severity lowered from low because the surrounding text warns against it.
Fix: human edits the `TELEGRAM_BOT_TOKEN=` line in place (`sudo nano <path>`, never a shellSuggested fix: Avoid sudo where possible. If it is required, explain why so the user can decide.
Checks run
- prompt-injection
- dangerous-commands
- credentials
- network
- file-access
- integrity
Checks skipped
- dependencies: No dependency manifest in the skill directory.
- gitleaks: gitleaks is not installed on this host.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 467k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 467k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 467k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 193k | Updated | Claude CodeCodex | Verified | Install |