Skillv1.0.1

ClawScan security

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

ReviewMar 15, 2026, 6:09 AM

Verdict: Review
Confidence: medium
Model: gpt-5-mini
Summary: The skill is internally consistent with building company intelligence but it silently depends on a local HTTP service (127.0.0.1:8010) that is not documented in the description or manifest, which is a potential surprise/risk for users.
Guidance: This skill expects a local service at http://127.0.0.1:8010/api/company/intel even though the manifest doesn't say so. Before installing or invoking it: (1) confirm you run and trust a local backend on port 8010 that implements this endpoint; (2) avoid sending sensitive credentials or private JD text unless you control and trust that service; (3) if you don't run such a service, the skill will fail harmlessly, but if you do run other local services bound to that port they could receive the POSTed data unintentionally; (4) ask the skill author for source code or a homepage to verify what the backend does, or run the skill in a sandboxed environment and test with non-sensitive sample inputs first.

Purpose & Capability: noteThe skill's stated purpose (build company intel) matches the actions in SKILL.md (POST to an /api/company/intel endpoint and parse structured results). However the skill does not declare or document that it requires a local backend running on 127.0.0.1:8010; that runtime dependency is implicit in the instructions and not mentioned in the description or registry metadata.
Instruction Scope: concernRuntime instructions instruct the agent to POST user-provided data to http://127.0.0.1:8010/api/company/intel and parse the response. The instructions do not read local files or env vars, but the unconditional network call to localhost can reach any service bound to that port on the user's machine. Because the skill doesn't document this dependency, a user could unknowingly route sensitive inputs to an unintended local service.
Install Mechanism: okNo install spec and no code files are included (instruction-only). This minimizes disk-write risk; there is nothing being downloaded or installed by the skill itself.
Credentials: okThe skill requests no environment variables, credentials, or config paths. The lack of credential requests is proportionate to its documented behavior.
Persistence & Privilege: okalways is false and the skill does not request persistent privileges or attempt to modify agent configuration. Autonomous invocation is allowed (platform default), which is expected for skills.