writing-tasks

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from local markdown files in the specs/ directory to automate the creation of GitHub issues.
  • Ingestion points: Specification files (.md) located in the project's specification root, as described in SKILL.md and decomposition.md.
  • Boundary markers: The instructions do not define explicit boundary markers or "ignore" instructions for the ingested content, leaving the agent susceptible to instructions embedded within the specification files.
  • Capability inventory: The skill utilizes Bash to execute GitHub CLI commands (gh issue create, gh issue edit) and uses Write/Edit tools to update specification frontmatter.
  • Sanitization: No sanitization, escaping, or validation of the extracted specification content is mandated before it is interpolated into shell command arguments or issue bodies.
  • [COMMAND_EXECUTION]: The skill makes extensive use of the Bash tool to interact with the GitHub CLI (gh). It dynamically constructs commands to list, view, create, and edit issues based on data parsed from local specification files.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 08:07 PM