common-context-optimization
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of technical documentation and strategies for managing large contexts. No malicious patterns such as prompt injection, persistence, or data exfiltration were identified in the instructions or reference files.
- [INDIRECT_PROMPT_INJECTION]: The skill describes techniques for processing and summarizing external data, such as tool outputs and logs. While this involves an ingestion surface for potentially untrusted data, the skill's primary function is to optimize token usage, and it does not possess any exploitable capabilities like network access or file system modification.
- Ingestion points: Raw tool outputs (logs, JSON lists) and chat history are identified as targets for optimization in
SKILL.md. - Boundary markers: No specific delimiters are defined for the summarized content.
- Capability inventory: No subprocess calls, network operations, or file-write capabilities are present in the skill.
- Sanitization: No specific sanitization or filtering logic is mentioned for the external data being summarized.
Audit Metadata