setup-chalk

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

• Third-party content exposure detected (high risk: 0.78). The skill’s runtime workflow reads multiple outsider-authored free-text files from the target repository (e.g., README.md, dependency manifests, CI/workflow YAML, and especially existing AGENTS.md/.cursorrules/CLAUDE.md) and then uses that content to generate .chalk/docs/* and chalk.json, which are fed back into the agent’s LLM context—this is an indirect prompt-injection path from repository text not authored by the operating user.