pp-gladly

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill's Auth Setup includes explicit examples that place credentials directly into shell commands and environment variables (e.g., export GLADLY_PASSWORD and gladly-pp-cli auth set-token "" ""), which encourages embedding secrets verbatim in agent output.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.75). The skill is a CLI that fetches/prints Gladly customer timeline/conversation data at runtime (e.g., via commands like conversations list/get and other read-only API reads), so the agent will ingest outsider-authored free text returned by the Gladly API into the LLM context through the CLI’s JSON output.