openspec-task-loop

Pass

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill includes a shell script, scripts/new_task_change.sh, which is used to generate project directories and markdown files. This script uses standard utilities such as mkdir and cat to perform local scaffolding within the project directory.
  • [PROMPT_INJECTION]: The skill processes task instructions from a local tasks.md file, which introduces a surface for indirect prompt injection if the file is populated from untrusted sources.
  • Ingestion points: Task identifiers and descriptions are read from tasks.md to define the scope of the agent's work loop.
  • Boundary markers: No specific delimiters or safety instructions are defined to separate task content from system instructions.
  • Capability inventory: The skill allows the execution of the local new_task_change.sh script and general file-system writes for project documentation.
  • Sanitization: Input from the task file is used directly in shell script arguments and file content without additional validation or sanitization.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 4, 2026, 10:59 PM