Back to skill
Skillv1.0.1
ClawScan security
Promociones de Viajes Argentina · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignFeb 19, 2026, 10:01 PM
- Verdict
- Benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- The skill is internally consistent: it only fetches a public promos endpoint and filters results locally; nothing requested or instructed is disproportionate to its stated purpose.
- Guidance
- This skill appears coherent and low-risk, but consider the following before installing: (1) it makes outbound requests to https://anduin.ferminrp.com — verify you trust that domain and its data/privacy practices, since responses come from an external host; (2) the instructions assume curl and jq are available on the agent runtime even though the manifest declares no required binaries—ensure those tools exist or adapt the instructions; (3) test the endpoint manually (curl the /api/v1/promos and inspect the JSON) to confirm the data/links are acceptable; (4) avoid sending any sensitive or personal data in queries since requests go to a third-party API; and (5) watch for future versions that might request credentials or broader permissions. If these points are acceptable, installing looks reasonable.
Review Dimensions
- Purpose & Capability
- okName/description (travel promos for Argentina) align with the runtime instructions: a single public API endpoint (https://anduin.ferminrp.com/api/v1/promos) is queried and results are filtered/ordered locally. No unrelated capabilities or credentials are requested.
- Instruction Scope
- noteSKILL.md confines actions to one GET endpoint and local filtering with jq; it does retry on network errors and describes precise output formatting. Minor mismatch: examples and workflow assume availability of curl and jq, but the skill does not declare required binaries. The instructions do not request reading local files, environment variables, or other system data.
- Install Mechanism
- noteThis is instruction-only (no install spec / no code written to disk), which is low risk. Note: the instructions rely on command-line tools (curl, jq) but the manifest lists no required binaries or install steps — this is a small consistency issue (runtime environment must provide those tools).
- Credentials
- okThe skill requests no environment variables or credentials and the API is documented as not requiring auth. There are no extraneous secrets or config paths requested.
- Persistence & Privilege
- okalways is false, the skill has no install-time persistence or system-wide modifications. It does not request permission to run always or to modify other skills or configs.