tldr-expert

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION] (SAFE): The skill utilizes standard command-line tools (repomix, gitingest) and MCP tools (llm-tldr) for their intended purposes. The examples provided are well-structured and do not involve shell piping to untrusted sources or high-privilege operations.
  • [DATA_EXPOSURE] (SAFE): The skill explicitly addresses data security by recommending the use of secretlint and configured ignore-lists to ensure that API keys, PII, and sensitive metadata are excluded from context bundles.
  • [INDIRECT_PROMPT_INJECTION] (LOW): As the skill involves ingesting and summarizing external codebases, there is a inherent surface for indirect prompt injection. However, the skill implements defensive patterns such as XML boundary markers and signature-only extraction to reduce the likelihood of executing instructions hidden in data.
  • [PROMPT_INJECTION] (SAFE): No patterns of system prompt extraction, safety filter bypass, or 'DAN-style' role-play were detected in the instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:22 PM