Skillv1.0.1

ClawScan security

Amazon Opportunity Discoverer · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

BenignApr 13, 2026, 12:37 PM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill's code, runtime instructions, and requested credential (APICLAW_API_KEY) are consistent with an Amazon product-opportunity scanner and do not request unrelated privileges.
Guidance: This skill appears coherent with its purpose: it needs only APICLAW_API_KEY and calls the APIClaw endpoints described. Before installing, verify you trust the source (review the repo/homepage), supply the API key via an environment variable rather than dropping it into a shared config.json in the skill directory, and be aware the included Python script will make network calls to https://api.apiclaw.io. If you want extra assurance, inspect the full scripts/apiclaw.py file for any unexpected endpoints or logging/transmission of your key, and consider creating a limited/monitoring API key (or usage alerts) on apiclaw.io to detect unexpected usage.

Purpose & Capability: okName/description promise (Amazon opportunity scanner) matches the provided script and SKILL.md which call APIClaw endpoints; the only required credential is APICLAW_API_KEY which is appropriate for the declared API usage.
Instruction Scope: noteSKILL.md restricts actions to calling APIClaw endpoints and running the included script; instructions do not request unrelated files or system credentials. Minor scope note: the script will also look for a local config.json beside scripts/ as an alternative source for the API key (documented in code), so credential placement can be either env var or that skill-local file.
Install Mechanism: okNo install spec or external downloads; this is an instruction-only skill with an included Python script. There is no remote install or archive extraction that would increase risk.
Credentials: noteOnly APICLAW_API_KEY is required and is the expected credential for the declared API. Caveat: the script supports reading an api_key from a config.json in the skill directory, which could expose the key if stored in a shared location—use the environment variable for least exposure.
Persistence & Privilege: okalways:false and default autonomy settings are used. The skill does not request persistent system-wide privileges or modify other skills; it only reads its own local config path (optional).