tidb-query-tuning

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

• Third-party content exposure detected (high risk: 0.90). The skill's AGENTS.md and SKILL.md explicitly instruct the agent to mine and search public GitHub issues (e.g., using scripts/generate_tidb_issue_experiences.py and the references/tidb-customer-planner-issues corpus) so the agent will fetch and interpret untrusted, user-generated content from GitHub as part of its workflow.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs running an issue-mining script and cloning/inspecting the pingcap/tidb GitHub repo (e.g., https://github.com/pingcap/tidb.git and references to https://github.com/pingcap/tidb/issues), which will fetch external GitHub content at runtime and can be injected into the agent’s local corpus and thus directly influence prompts/behavior.