Back to skill
Skillv1.0.0
ClawScan security
Lead Finder Outreach · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignApr 9, 2026, 10:34 AM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- The skill's requests and instructions are consistent with its stated purpose (building defensible prospect shortlists from public sources); it is instruction-only, asks for no credentials, and contains no install steps.
- Guidance
- This skill appears coherent and limited to public-source research, but exercise normal caution before using outputs for outreach: 1) Verify each contact path and URL the skill cites — LLMs can hallucinate or misattribute details even when instructed not to. 2) Do not rely on this pack for private-data enrichment or for harvesting emails that are not publicly posted. 3) Ensure any outreach complies with applicable laws (CAN-SPAM, GDPR, local rules) and with the target site's terms of use. 4) If a future version requests API keys, account tokens, or installs tools for scraping or bulk emailing, treat that as a red flag unless you clearly understand and approve the external service and its permissions. 5) Prefer manual or automated verification of at least a sample of entries before importing into a CRM or sending first-touch outreach.
Review Dimensions
- Purpose & Capability
- okName/description (finding agency-ready prospects, public contact paths, 50-company shortlists) align with the skill's contents. The skill asks for no unrelated binaries, credentials, or config paths and does not attempt to access resources beyond public web sources, which is proportionate for this purpose.
- Instruction Scope
- noteSKILL.md and referenced docs narrowly define a public-sources-only workflow, require minimal input (niche, location, fit criteria), and explicitly forbid fabricating private data or mass-scraping. One practical risk to note: language models can hallucinate contact details or overstate confidence even when instructions forbid fabrication — the skill properly documents verification steps, but outputs should be spot-checked against the cited public URLs before use.
- Install Mechanism
- okNo install spec and no code files — instruction-only skill. Nothing is downloaded or written to disk by the skill bundle itself, which minimizes supply-chain and execution risk.
- Credentials
- okThe skill declares no required environment variables, no primary credential, and no config paths. That matches the documented behavior of researching public web sources and producing formatted output.
- Persistence & Privilege
- okalways is false and the skill does not request persistent system-level privileges or modifications. Default autonomous invocation is allowed (platform default) but not granted extra power by the skill itself.