build-smart-contracts

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

• Third-party content exposure detected (high risk: 0.85). The required workflow includes “Retrieve canonical examples” from public GitHub repositories (e.g., algorandfoundation/devportal-code-examples, algorandfoundation/puya-ts, algorandfoundation/algokit-typescript-template) and then “Generate code adapting examples,” which implies the agent ingests outsider-authored free-form source text from those repos into the LLM context at runtime.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly requires retrieving canonical examples from GitHub at runtime (e.g. repos like https://github.com/algorandfoundation/devportal-code-examples and https://github.com/algorandfoundation/puya-ts) and uses those fetched examples to drive code generation and tests, meaning external content fetched during runtime directly controls the agent's outputs and is a required dependency.