vestige-context

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes the vestige command-line utility to fetch project metadata. This includes parameters for token budgeting and section limits, which are derived from user intent. The command execution is the primary mechanism for the skill's functionality.
  • [PROMPT_INJECTION]: The skill is subject to indirect prompt injection because it ingests external data (project summaries, decisions, open questions) into the agent's context. Malicious content within the project's memory store could attempt to influence the agent's behavior.
  • Ingestion points: Data returned by the vestige context --json command is directly formatted for the agent's use.
  • Boundary markers: No specific delimiters or 'ignore' instructions are used to wrap the retrieved project context.
  • Capability inventory: The skill allows for command execution (vestige context) and subsequent information retrieval (vestige-show).
  • Sanitization: The skill does not explicitly mention sanitizing or validating the content retrieved from the project context pack before it is presented to the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 09:23 AM
Security Audit — agent-trust-hub — vestige-context