settings-and-tuning

Pass

Audited by Gen Agent Trust Hub on Apr 4, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses Bash to manage the bot process, specifically using pkill to stop existing instances and bun to restart the service. These operations are scoped to the project directory and specific process patterns.
  • [DATA_EXFILTRATION]: The skill reads configuration data from ~/.claude/automation/claude-telegram-sync/mise.toml. While this file contains environment variables, the skill architecture correctly separates secrets into a gitignored .mise.local.toml file which is not targeted for modification by this skill.
  • [EXTERNAL_DOWNLOADS]: Mentions brew install mise as a prerequisite. This is a reference to a well-known package manager and utility, used here for legitimate environment management.
  • [PROMPT_INJECTION]: Includes a 'Self-Evolving' instruction that encourages the agent to update the skill's own documentation and instructions based on usage. This is an intended functional pattern for the skill's purpose and does not contain malicious bypass attempts.
  • [SAFE]: The skill follows established best practices for configuration management by centralizing settings in a single source of truth (mise.toml) and providing clear validation steps and troubleshooting guidance.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 4, 2026, 09:51 AM