Skillv2.1.1

ClawScan security

Cpo Advisor · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

BenignMar 10, 2026, 10:04 PM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill's files and runtime instructions align with its stated purpose (CPO/advisory work) and request no extraneous credentials or installs; minor metadata omissions should be noted before running.
Guidance: This skill appears coherent and focused on product strategy. Before running: (1) ensure Python is available (SKILL.md uses 'python' but the metadata didn't declare it); (2) inspect the two scripts (pmf_scorer.py, portfolio_analyzer.py) yourself — they operate on any JSON/data you supply, so avoid feeding sensitive PII unless you trust the environment; (3) run the scripts in a controlled environment (local dev machine or sandbox) if you have any doubt; (4) if you plan to run them inside an agent, confirm the agent won't inadvertently upload your data elsewhere — the skill's files show no network calls, but it's good practice to validate in your runtime. If you need the skill to declare runtime dependencies, ask the author to list Python (and a recommended version) in the metadata.

Review Dimensions

Purpose & Capability: noteName/description (product leadership, PMF, portfolio strategy) match the included reference docs and two analysis scripts. One minor inconsistency: SKILL.md expects you to run 'python scripts/…' but the registry metadata did not declare a required 'python' binary; this is an omission (not evidence of malicious intent) but should be addressed.
Instruction Scope: okSKILL.md instructs the agent/user to run local Python scripts and consult included reference docs. The instructions stay within the product-strategy/PMF/portfolio domain and only ask for product analytics input (e.g., retention cohorts) that are appropriate for the stated purpose. There are no instructions to read unrelated system files or send data to external endpoints.
Install Mechanism: okThere is no install spec (instruction-only behavior) and included code files are local. No network downloads, package installs, or archive extraction are present in the manifest. The only runtime dependency is Python, which should be present on the host.
Credentials: okThe skill does not request any environment variables, credentials, or config paths. The scripts operate on user-provided JSON/data and local inputs only; no secrets are required or requested.
Persistence & Privilege: okThe skill is not always-enabled and does not request elevated persistence or modify other skills/system settings. It is user-invocable and may be invoked autonomously by the agent per platform defaults, which is expected for skills of this type.