improve-coverage

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

• Third-party content exposure detected (medium risk: 0.30). The required workflow runs tools/run-cov <target> -r llm, then has the LLM read the generated coverage report and source/test files named in that report, meaning outsider-submitted target/path strings can cause the agent to ingest report text derived from the repository’s analyzed code/tests.