gsd-resume-work

Pass

Audited by Gen Agent Trust Hub on Jun 21, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as an integration layer for the 'Get Shit Done' (GSD) framework, providing mapping rules between framework-specific commands (AskUserQuestion, Task) and the agent's native tools (request_user_input, spawn_agent).
  • [DATA_EXFILTRATION]: The skill references absolute file paths within the author's home directory (/home/delorenj/code/bhappy/) to load workflows and project state. These operations are local to the workstation and intended for context restoration, showing no signs of unauthorized data access or exfiltration.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from local project files to reconstruct session state, presenting a standard surface for indirect instructions.
  • Ingestion points: Processes STATE.md, .continue-here, and CONTEXT.md files from the local project path.
  • Boundary markers: Not present in this SKILL.md file; logic is delegated to the resume-project.md workflow.
  • Capability inventory: Can spawn sub-agents (spawn_agent) and request user input (request_user_input).
  • Sanitization: None explicitly defined in the instruction set.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 21, 2026, 04:26 AM
Security Audit — agent-trust-hub — gsd-resume-work