skills/itumulak/kahanas/dev-document/Gen Agent Trust Hub

dev-document

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses Bash to execute git for history analysis and branch management, and gh (GitHub CLI) to create and update pull requests.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface by ingesting external, untrusted data to generate documentation.
  • Ingestion points: Data is read from git log, git diff, decision-log.md, build-plan.md, and incident facts provided via the AskUserQuestion tool.
  • Boundary markers: Uses predefined Markdown templates in the templates/ directory to structure generated content.
  • Capability inventory: Includes the ability to push code to remote repositories, manage GitHub pull requests via the API, and write to the local filesystem (CHANGELOG.md, .konteksto/).
  • Sanitization: Employs safety measures such as writing large text blocks to temporary files (--body-file) to avoid shell injection and includes explicit instructions to identify and mask secrets found in source material.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 05:43 AM