loop-finder

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt requires reading local chat session files and extracting & quoting user message text as "evidence," which risks exposing API keys, tokens, or passwords present in those messages (i.e., outputting secrets verbatim).

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.80). The prompt instructs the agent to write/append persistent agent system-instruction files in user config directories (e.g., ~/.pi/agent/AGENTS.md, ~/.codex/AGENTS.md, ~/.claude/CLAUDE.md), which modifies the machine's state and can alter assistant behavior — a persistent configuration change risk even though it doesn't request sudo or create users.