task-planning

Pass

Audited by Gen Agent Trust Hub on Oct 29, 2026

Risk Level: SAFEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The script tests/run-validation.py uses exec() and compile() to load and execute the contents of the internal utility script scripts/validate-output.py.
  • Evidence: exec(compile(validator_source, str(SKILL_DIR / "scripts" / "validate-output.py"), "exec"), validator_namespace) in tests/run-validation.py.
  • Context: This pattern is used as a lightweight mechanism to import and execute testing logic from a sibling file within the skill's own package without formal module installation. It targets a static local file path.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted user input in the form of project specifications and requirements, which defines an attack surface for indirect prompt injection.
  • Ingestion points: User-provided project specifications, features, or migration briefs described in SKILL.md.
  • Boundary markers: Absent; the instructions do not specify the use of delimiters or specific warnings to ignore embedded instructions within user-provided specifications.
  • Capability inventory: Execution of internal Python scripts scripts/validate-output.py and tests/run-validation.py via shell commands; file reading of bundled assets and rules.
  • Sanitization: Absent; user input is processed as raw text without explicit sanitization or validation of its content prior to processing by the model.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 29, 2026, 03:18 PM