executar-task

Pass

Audited by Gen Agent Trust Hub on Apr 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection. It ingest instructions from external markdown files located in the project's task directory.
  • Ingestion points: Processes content from ./tasks/prd-[feature-slug]/[num]_task.md, prd.md, and techspec.md.
  • Boundary markers: The instructions do not specify any delimiters or safety markers to differentiate between the skill's logic and the data read from these files.
  • Capability inventory: The agent has the capability to write code and execute shell commands (to run tests).
  • Sanitization: There is no evidence of sanitization or validation logic for the content read from the task files.
  • [COMMAND_EXECUTION]: The skill requires the agent to 'Create and run all task tests', which inherently involves shell command execution. When paired with the ingestion of external task definitions, this presents a risk where a malicious task file could influence the agent to execute harmful commands during the testing phase.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 5, 2026, 03:23 AM
Security Audit — agent-trust-hub — executar-task