to-maestro-tasks
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Node.js script to synchronize the status of generated tasks.
- Evidence: Step 6 invokes
node "${CLAUDE_PLUGIN_ROOT}/scripts/maestro-task-status.cjs" syncto parse task files and update a local JSON status tracker. - Context: The script is located within the plugin's own root directory (
CLAUDE_PLUGIN_ROOT), representing internal utility functionality rather than an external or untrusted download. - [INDIRECT_PROMPT_INJECTION]: The skill generates task prompts based on user-supplied project specifications, which could potentially include instructions that influence downstream agent actions.
- Ingestion points: User-provided plans, specifications, design documents, or PRDs processed during the drafting phase.
- Boundary markers: Uses a structured Markdown template (
task-file-template) to encapsulate generated content within specific headings like 'What to build' and 'Acceptance criteria'. - Capability inventory: Writing files to the local
.claude/maestro-tasks/directory and executing a bundled Node.js maintenance script. - Sanitization: No explicit sanitization or filtering of user-provided content is mentioned before it is interpolated into the task file templates.
Audit Metadata