skills/bcms/ai/bcms-content/Gen Agent Trust Hub

bcms-content

Pass

Audited by Gen Agent Trust Hub on Aug 4, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONCOMMAND_EXECUTION
Full Analysis
  • [UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: The skill utilizes the official @thebcms/client SDK to facilitate communication with the BCMS API.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: Authenticates using the BCMS_API_KEY environment variable, following secure practices for credential management by avoiding hardcoded secrets.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: Includes a media upload command that reads files from the local filesystem and transmits them to the official BCMS media library, which is consistent with the skill's primary purpose.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it retrieves content from an external API and has write capabilities. This represents a standard risk for content management tools and no specific malicious triggers were detected.
  • Ingestion points: cli/bcms.mjs retrieves content via the createEntry, updateEntry, and listEntries functions.
  • Boundary markers: No explicit boundary markers are present in the JSON data returned from the API.
  • Capability inventory: The skill can create, modify, and delete entries on the BCMS platform, and can read local files for media uploads.
  • Sanitization: No content filtering or escaping is applied to the data fetched from the API.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 4, 2026, 08:37 AM
Security Audit — agent-trust-hub — bcms-content