skills/laststance/skills/save/Gen Agent Trust Hub

save

Pass

Audited by Gen Agent Trust Hub on May 1, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill relies exclusively on specialized MCP memory tools (write_memory, read_memory, list_memories) to manage session persistence. No suspicious network operations or shell commands were identified across the analyzed files.
  • [PROMPT_INJECTION]: The skill possesses an Indirect Prompt Injection surface, which is inherent to its functionality of summarizing project files and activity.
  • Ingestion points: In SKILL.md Phase 1, the agent is instructed to review created or modified files and session decisions to identify content for storage.
  • Boundary markers: The instructions lack specific boundary markers or warnings to ignore embedded instructions within the processed data.
  • Capability inventory: The skill uses write_memory and read_memory to store and retrieve analyzed context.
  • Sanitization: No explicit sanitization or filtering logic is provided for the data gathered from the workspace before it is written to memory.
Audit Metadata
Risk Level
SAFE
Analyzed
May 1, 2026, 12:54 AM