openshift-docs

Pass

Audited by Gen Agent Trust Hub on Jul 28, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The Makefile includes a target to clone the openshift/openshift-docs repository from GitHub. This is an official source belonging to the vendor, used to populate the skill's documentation content.
  • [COMMAND_EXECUTION]: The convert.py script, provided for documentation maintenance, uses the subprocess.run method to execute asciidoctor and pandoc system utilities. These commands are constructed using secure list-based arguments rather than shell strings, mitigating command injection risks.
  • [PROMPT_INJECTION]: The skill provides an interface for the agent to ingest and process a large volume of external documentation files. While the source is reputable, the capability to read and act on external text constitutes a standard indirect prompt injection surface for RAG-based systems. Ingestion points include all files under the docs/ directory. Boundary markers and explicit sanitization are absent, however, the agent's capabilities are limited to the read-only tools cat, grep, find, and ls within the local documentation directory.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 28, 2026, 11:08 PM
Security Audit — agent-trust-hub — openshift-docs