Back to skill
Skillv1.0.0
ClawScan security
Nex Skillmon · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignApr 5, 2026, 4:51 AM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- The skill's code, runtime instructions, and requested environment access are consistent with its stated purpose (scanning a skills directory, checking ClawHub for flags/updates, and tracking local usage/costs).
- Guidance
- This package appears to do what it says: it scans a local skills directory, records metadata and usage in ~/.nex-skillmon, computes SHA256 hashes of skill files, and calls the configured ClawHub API to check for flags/updates. Before installing: (1) verify SKILLS_BASE_DIR points only to the directory you intend scanned (avoid pointing it to sensitive system paths), (2) review the ClawHub endpoint you configure (CLAWHUB_API_URL) if you don't want skill names sent externally — you can disable remote checks by setting CLAWHUB_API_URL to empty as documented, (3) note the installer will create ~/.nex-skillmon and add a user-level executable link (~/.local/bin), and (4) review the provided source if you want to audit what is hashed/posted. If you want to limit exposure, run the tool in a contained environment (VM/container) or review/modify scanner.check_for_updates and check_security_flags to point to a trusted endpoint or disable them.
Review Dimensions
- Purpose & Capability
- okName/description (health monitoring, cost tracking, security checks) match the files and behavior: scanner reads SKILL.md files under a skills directory, computes hashes, calls a ClawHub API for updates/flags, and stores local telemetry in ~/.nex-skillmon. Required binaries (python3) and the declared env vars (CLAWHUB_API_URL, SKILLS_BASE_DIR) are appropriate for the stated functionality.
- Instruction Scope
- noteRuntime instructions and CLI commands are focused on discovery, health checks, security scans, and cost reporting. The scanner will read all files under SKILLS_BASE_DIR and compute SHA256 hashes; it also makes outbound HTTP requests to CLAWHUB_API_URL for update/flag checks. These behaviors are expected for a monitoring tool, but users should be aware the skill will read the full contents of skill directories (file hashing) and send skill names to the configured ClawHub endpoint.
- Install Mechanism
- okNo external downloads or package installs are performed. setup.sh initializes a local SQLite DB, creates a ~/.nex-skillmon data directory and logs, and creates a symlink/wrapper in ~/.local/bin (or a Windows batch wrapper). The installer is local and idempotent; no network fetches or archives are pulled during setup.
- Credentials
- okOnly CLAWHUB_API_URL and SKILLS_BASE_DIR are declared as required env vars; these are proportionate to checking a remote skill index and scanning a local skills directory. The skill does read other optional env vars (CURRENCY, LOG_LEVEL) but these are non-sensitive configuration values. The package does not request secret tokens, cloud keys, or unrelated credentials.
- Persistence & Privilege
- noteThe skill stores persistent state under the user's home (~/.nex-skillmon) and creates a user-level symlink/wrapper in ~/.local/bin. always is false. This level of persistence is typical for a CLI monitoring tool; it does not modify other skills or system-wide settings beyond adding a user bin entry.