prompts-chat

Pass

Audited by Gen Agent Trust Hub on Jul 28, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill clones the prompts.chat repository from GitHub to provide access to curated prompt templates. This is an expected behavior for a prompt library tool targeting a well-known community resource.
  • [COMMAND_EXECUTION]: The discover-prompts.sh script executes standard system commands including git, curl, jq, find, and grep. These are used to maintain a local cache and perform search operations on the fetched markdown files.
  • [PROMPT_INJECTION]: Because the skill ingests content from an external source (the prompt library), it creates an indirect prompt injection surface. The skill instructions mitigate this by including steps for users and agents to evaluate, select, and adapt the prompts to their specific context before use.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 28, 2026, 01:51 PM
Security Audit — agent-trust-hub — prompts-chat