session-handoff

Pass

Audited by Gen Agent Trust Hub on May 8, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface. The skill facilitates the ingestion of untrusted data through its handoff mechanism. 1. Ingestion points: The references/load.md file defines the workflow for reading session snapshots from .artifacts/.session-handoff.md into the active context. 2. Boundary markers: The instructions lack explicit delimiters or 'ignore embedded instructions' warnings for the loaded data. 3. Capability inventory: The agent possesses standard tool capabilities (e.g., shell access, file writes) that could be exploited if malicious instructions are loaded into the conversation context. 4. Sanitization: There is no validation or sanitization of the snapshot content before it is processed by the agent.
  • [SAFE]: Local Operation. The skill manages session state using a local file and does not request network access, perform external downloads, or access sensitive system paths like credentials or SSH keys.
Audit Metadata
Risk Level
SAFE
Analyzed
May 8, 2026, 07:05 PM