rhdh-catalog-index

Pass

Audited by Gen Agent Trust Hub on Apr 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes several command-line tools including skopeo, tar, jq, and python3 to manage OCI images and extract their contents. These operations are restricted to local temporary directories and used for inspection purposes. \n- [EXTERNAL_DOWNLOADS]: The skill downloads OCI images and metadata from quay.io and ghcr.io. These are well-known container registries used for distributing official software components. \n- [DATA_EXFILTRATION]: No evidence of unauthorized data transmission was detected. The skill uses network operations to fetch metadata and images from reputable sources. \n- [PROMPT_INJECTION]: The skill processes content from external OCI images, which introduces a potential attack surface for indirect prompt injection. \n
  • Ingestion points: Data is ingested from the local package.json file and remote OCI image artifacts (layers and annotations). \n
  • Boundary markers: The skill does not define specific delimiters or include instructions for the agent to ignore potential instructions embedded within the extracted YAML or metadata. \n
  • Capability inventory: The skill allows the execution of system commands (skopeo, tar, python3) and filesystem writes to temporary directories. \n
  • Sanitization: Structural validation is provided via JSON and YAML parsing, but the skill lacks semantic sanitization of the values retrieved from external registries.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 27, 2026, 02:14 PM