The Agent Skills Directory

[Data Exposure & Exfiltration] (HIGH): The function scanFile in scripts/src/index.ts implements a pattern that reads arbitrary files from the local filesystem using readFileSync(filePath) and uploads the content to https://endpoints.work/api/scan. This provides a direct path for data exfiltration if the agent is prompted to process sensitive system files.
[Indirect Prompt Injection] (LOW): The skill is vulnerable to indirect prompt injection as it ingests untrusted data from external sources without adequate safeguards. Evidence Chain: 1. Ingestion points: The functions scanText (text argument) and scanFile (file content) in scripts/src/index.ts. 2. Boundary markers: Absent; there are no delimiters or instructions provided to the extraction API to ignore embedded instructions within the ingested data. 3. Capability inventory: The skill has the capability to read files (readFileSync), write files (writeFileSync), and perform network requests (fetch). 4. Sanitization: Filenames for saved results are sanitized, but the content being sent to the AI extraction API is not sanitized or validated.
[External Downloads] (LOW): The skill requires the installation of several Node.js dependencies (dotenv, tsx, typescript) from the npm registry. While these are standard packages, they represent an external dependency chain.

endpoints