rlm-long-context

Pass

Audited by Gen Agent Trust Hub on Jul 12, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill's core functionality relies on executing local Python scripts (orchestrate.py, rank_chunks.py, semantic_chunk.py, and codebase_concat.py) via the command line. These scripts perform file operations within the project's workspace to prepare and process data for the AI agent.
  • [DYNAMIC_EXECUTION]: The file scripts/orchestrate.py uses importlib.import_module() to dynamically load internal helper scripts (cache_manager and rank_chunks). This pattern is used to support both package-based and standalone execution of the orchestration logic. The modules being loaded are local to the skill's directory and the module names are hardcoded.
  • [DATA_EXPOSURE]: The scripts/codebase_concat.py script automates the reading and aggregation of multiple files from a local directory into a single text file for context processing. While this involves broad read access to the local codebase, it includes standard security practices such as excluding hidden directories (e.g., .git) and dependency folders (e.g., node_modules) to prevent unintended data exposure.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 12, 2026, 02:01 AM
Security Audit — agent-trust-hub — rlm-long-context