agent-startup-and-context-ingestion
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill functions as a documentation and methodology guide for agent startup and context management. It describes a conceptual framework for how agents become useful by ingesting roles, specs, and checklists without implementing any dangerous operations.
- [COMMAND_EXECUTION]: The skill references the use of
rig whoamiandrig context get. These are utility CLI commands used to resolve the agent's identity and retrieve project-specific metadata from the local environment. - [INDIRECT_PROMPT_INJECTION]: The skill defines a process for ingesting external data from sources such as
project.yaml,mission.yaml, and session transcripts. While this represents a data ingestion surface, the skill provides specific precedence rules and address maps to resolve authority, which is a standard part of its architectural purpose.
Audit Metadata