structured-context-compressor

Pass

Audited by Gen Agent Trust Hub on Apr 1, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [SAFE]: The skill's primary function is to provide a template and prompt for summarizing conversation history. This is a standard utility function for improving LLM context management.
  • [COMMAND_EXECUTION]: The skill includes a helper script (scripts/render_template.py) which is designed to be executed to print a template. Analysis of the source code confirms it only performs standard output operations and contains no malicious logic, network access, or file system modifications.
  • [PROMPT_INJECTION]: The skill includes a prompt template (references/prompt-template.md) that guides the agent's summarization process. These instructions are functional and do not attempt to bypass safety filters or override system-level constraints. Although the skill processes conversation history (indirect prompt injection surface), it lacks the high-privilege capabilities necessary to exploit such a vector.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 1, 2026, 11:23 AM