session-save

Pass

Audited by Gen Agent Trust Hub on Oct 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill captures the current session state, including potentially untrusted task descriptions or code snippets, and writes them to a local file (session-notes.md). This creates a vulnerability surface where malicious instructions could be persisted and later re-processed by the agent.
  • Ingestion points: The skill ingests the entire current agent context (tasks, completed items, decisions) to generate the summary.
  • Boundary markers: The generated file structure lacks explicit boundary markers or instructions telling the agent to ignore embedded commands within the notes.
  • Capability inventory: The skill utilizes file-writing capabilities to create and overwrite session-notes.md in the project root.
  • Sanitization: There is no evidence of sanitization or escaping of the session content before it is written to the file.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 15, 2026, 04:41 PM