synthesis-okf

Pass

Audited by Gen Agent Trust Hub on Aug 14, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/okf_convert.py executes the git log command via subprocess.run to retrieve the last commit timestamp for markdown files. The implementation follows security best practices by using a list of arguments and the -- separator to prevent argument injection from file paths.
  • [INDIRECT_PROMPT_INJECTION]: The skill's scripts ingest untrusted markdown files to extract titles and descriptions (Category 8 surface). The scripts use yaml.safe_load() for parsing frontmatter and do not interpret the content as instructions, minimizing the risk of indirect injection during the conversion process.
  • [DATA_EXPOSURE]: The tools read local markdown files and repository configuration files (.agents/knowledge-base.yaml). This file access is restricted to the local filesystem and is necessary for the skill's validation and conversion functions; no network-based data exfiltration was detected.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 14, 2026, 05:31 PM
Security Audit — agent-trust-hub — synthesis-okf