specstory-yak

Pass

Audited by Gen Agent Trust Hub on Oct 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external session history files to generate a report that the AI is instructed to summarize.
  • Ingestion points: History files located in .specstory/history/ are read by the main analysis script in scripts/analyze.py.
  • Boundary markers: The script output lacks explicit delimiters or warnings to the AI to treat the summarized history content as untrusted data.
  • Capability inventory: The skill has the capability to execute shell commands (git blame) and write report files to the filesystem.
  • Sanitization: Content is truncated in scripts/lib/parser.py but is not filtered for potential prompt injection sequences that might be embedded in the coding session history.
  • [COMMAND_EXECUTION]: The skill executes the git blame command to identify authors for the report leaderboard.
  • Evidence is found in scripts/lib/utils.py using subprocess.run.
  • The implementation uses a list of arguments rather than a shell string, which is a secure practice to prevent shell injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 6, 2026, 04:22 PM