project-new

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to create directories and write template files (project.md) to the local file system under ~/.config/finance-skills/projects/. This is a standard administrative operation for project management and does not involve arbitrary or dangerous command execution.
  • [DATA_EXPOSURE]: The skill follows security best practices by explicitly stating that project directories must not store API credentials, which are isolated in a separate configuration path. It uses the standard ~/.config/ directory for application data.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses an ingestion surface for untrusted data as it takes user-supplied strings (project names, owners, descriptions) and interpolates them into markdown templates.
  • Ingestion points: User inputs collected during Step 3 and Step 5 of the interaction flow.
  • Boundary markers: The markdown template does not currently define specific delimiters or warnings to ignore embedded instructions in the generated file.
  • Capability inventory: The agent is authorized to create directories and write files to the project space.
  • Sanitization: No specific sanitization or validation rules are provided for the user-supplied strings before they are written to disk. While this presents a minor theoretical surface for injection if the generated files are later parsed by an AI, it is inherent to the skill's intended purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 02:26 AM
Security Audit — agent-trust-hub — project-new