ns-pm-delivery-schedule

Pass

Audited by Gen Agent Trust Hub on Aug 8, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill is designed to run a local Python script (../ns-project-manager/scripts/pert_montecarlo.py) to perform Monte Carlo simulations. This script is a sibling component belonging to the same developer ecosystem. The execution path is relative, allowing for variations in the skill's installation directory.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because it ingests untrusted data from the user via an intake form (references/intake.md).
  • Ingestion points: The agent collects project phase IDs, names, and numerical estimates from user prompts.
  • Boundary markers: The instructions do not define delimiters or warnings to ignore instructions embedded in the user-provided text.
  • Capability inventory: The skill possesses the capability to write multiple files (05-cronograma-tres-cenarios.md, pert-config-*.json, roadmap.md) and execute shell commands via python3.
  • Sanitization: There is no evidence of escaping or validation of user-provided content before it is interpolated into the configuration files or the final project roadmap.
  • [FILE_SYSTEM_OPERATIONS]: The skill performs file system writes to persist project documentation and configuration files. It uses a discovery mechanism (../ns-harness/references/harness-discovery.md) to resolve the root directory, which is standard behavior for this suite of project management tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 8, 2026, 09:08 PM
Security Audit — agent-trust-hub — ns-pm-delivery-schedule