Skillv1.0.2

ClawScan security

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

Scanner verdict

SuspiciousApr 30, 2026, 1:54 AM

Verdict: suspicious
Confidence: medium
Model: gpt-5-mini
Summary: The skill's instructions claim autonomous scraping and auto-submitting proposals but the package provides no code, no install, and requests no credentials — several things don't line up and warrant caution.
Guidance: Do not install or enable autonomous execution yet. This skill is incomplete and ambiguous: it claims to auto-scrape and auto-submit proposals but provides no code, no install, and asks for no credentials. Before proceeding, ask the publisher for: (1) the actual scripts/source code (so you can review what will run), (2) a clear explanation of how authentication to FreeLanceDAO/CryptoGigs/OpenCollab is handled (where tokens are stored and how consent is obtained), and (3) a safe dry-run mode and explicit confirmation step before any proposals are submitted. If you want similar functionality, prefer a version that bundles or links to verifiable code (e.g., GitHub repo with releases), documents required credentials and scopes, and gives you manual approval before submissions. If you must try it, run only in a restricted/test environment, require dry-run mode, and never grant it real account credentials until you've audited the code.

Review Dimensions

Purpose & Capability: concernThe skill claims to find gigs, score them, auto-submit proposals, and run twice daily, but the bundle is instruction-only and contains no scripts, no install, and requests no credentials. Autonomous submission and profile updates normally require account authentication and code to run; those are missing here, so the declared capability is inconsistent with what is provided.
Instruction Scope: concernThe SKILL.md instructs scraping multiple third‑party sites and auto‑submitting personalized proposals (writing earnings_log.json and building profiles). It does not describe how credentials, rate limits, or consent are handled, nor does it require or reference any local files that would contain account auth. It also refers to python scripts (scripts/gig_hunter.py, scripts/build_cryptogig_profile.py) that are not included in the package — the runtime instructions therefore assume external code or access to the user's environment in ways not specified.
Install Mechanism: concernThere is no install spec (instruction-only), which is low-risk in itself, but the SKILL.md expects runnable scripts that are not bundled. That mismatch means the skill either expects preexisting local tooling (not declared) or is incomplete — both are problematic for safely running autonomous actions.
Credentials: concernNo environment variables, credentials, or config paths are declared, yet the described functionality (account profile creation, proposal submission, escrow checks) would normally require platform credentials and possibly API keys. The absence of requested credentials is disproportionate and unclear: where/how would the skill authenticate to these platforms?
Persistence & Privilege: okalways is false (not force-included) and model invocation is allowed by default. That autonomy is expected for skills of this type, but combined with the other concerns (auto-submission without clear credential handling or included code) it increases risk. The skill does not request elevated persistent privileges.