skills/endlessblink/flow-state/task/Gen Agent Trust Hub

task

Pass

Audited by Gen Agent Trust Hub on Apr 22, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Executes a local Node.js script located at ./scripts/utils/get-next-task-id.cjs to manage task identifiers.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted data from multiple sources. Ingestion points: Reads project status from docs/MASTER_PLAN.md and accepts multi-part user input via the AskUserQuestion tool. Boundary markers: None present; the skill lacks delimiters or instructions to ignore potential commands embedded in the project files or user responses. Capability inventory: The skill can execute local system commands via node and modify project files (docs/MASTER_PLAN.md). Sanitization: None present; there is no evidence of input validation or escaping before data is written to the file system or processed.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 22, 2026, 01:14 AM
Security Audit — agent-trust-hub — task