tiangong-kb-esg-search

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill explicitly accepts and prioritizes inline secret fields (esg_api_key / api_key) in forwarded JSON input, which can require the LLM to include secret values verbatim in generated requests rather than only using environment- or CLI-managed auth.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The wrapper/script defaults to running "npx @tiangong-ai/cli@latest", which fetches and executes remote npm package code at runtime and is relied on as the required CLI entrypoint (scripts/esg_search.sh sets CLI_COMMAND to this and SKILL.md documents it).