Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

Project Bootstrap

v1.0.0

Bootstrap a multi-agent software project from idea to running CI/CD. Use when starting a new project that needs agent team design, task management, GitHub re...

0· 85·0 current·0 all-time
byBingji Guo@ckouder
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Suspicious
high confidence
Purpose & Capability
The skill claims to bootstrap a multi-agent project and includes a taskboard CLI, CI/CD templates, and detailed orchestration instructions — the included files (SKILL.md, CI/CD templates, taskboard script) are coherent with that purpose. However, the SKILL metadata lists no required binaries or env vars even though the instructions and script assume tools (gh, Python) and services (GitHub, Discord).
!
Instruction Scope
Runtime instructions tell the agent to run gh CLI commands, create GitHub rulesets, add repo secrets, and configure Discord webhooks. The SKILL.md and the included script both perform network operations (GitHub API calls, webhook use) and write/read local files (taskboard.json, openclaw.json). These actions are within the skill's scope, but the instructions grant the agent the ability to create repos, set secrets, and post webhooks — high-impact operations that should be explicitly noted and approved by the user or operator.
Install Mechanism
No external install/downloads are performed; this is instruction-heavy and ships a single Python CLI script. No remote code is fetched by an installer during skill installation.
!
Credentials
The registry metadata declares no required environment variables, but the taskboard script expects an optional GITHUB_TOKEN (or custom token env name) and the CI/CD templates/instructions rely on a DISCORD_WEBHOOK secret being created in GitHub. The skill will operate differently depending on those secrets; the omission in the declared requirements is an inconsistency and means the agent or user might be asked to provide or create credentials at runtime without prior notice.
Persistence & Privilege
The skill is not always-enabled and does not request elevated platform privileges. It writes to local files (taskboard.json, cache file) and to repo settings via GitHub API when a token is provided, which are expected behaviors for this functionality.
What to consider before installing
This skill appears to do what it claims (set up agent roles, a local taskboard CLI, GitHub repo and Actions templates, and Discord notifications), but there are important inconsistencies and high-impact steps you should check before installing: - Required tools: The SKILL.md tells you to run gh and the code is a Python CLI. Ensure the host has GitHub CLI (gh) and python3 available; the skill metadata did not list these as required. - Secrets and tokens: The taskboard script will use a GITHUB_TOKEN (or a token env you configure) to sync with GitHub; the CI templates expect a GitHub secret named DISCORD_WEBHOOK for Discord notifications. Only grant the minimum scopes needed (e.g., repo/issues plus workflow and secrets as required) and review who has access to those tokens. - Review code & endpoints: scripts/taskboard.py makes HTTP requests to api.github.com and writes local files (taskboard.json). Inspect the script yourself to confirm endpoints and behavior are acceptable for your environment. - High-impact operations: The workflow includes creating repos, configuring branch protection rules, and adding repo secrets. These can modify org/repo settings — run them with an account that has appropriate permissions and consider testing in a throwaway repo first. - Audit templates: Branch protection JSON and Actions reference specific contexts and status checks (e.g., 'test' status). Verify these match the actual workflows in your repo before applying to main (branch protection with misconfigured checks can block merges). If you want to proceed: run the skill in a controlled/test repository or sandbox, provide least-privilege tokens, and verify/disallow any automated agent actions that create or modify GitHub org-level resources without explicit human approval.

Like a lobster shell, security has layers — review code before you run it.

latestvk978nq3rasyhp3vhwx5fsm323n836czg

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Comments