The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the local project environment, which could potentially contain malicious instructions intended to influence the agent's output.
Ingestion points: The agent is instructed to read .pythia/conventions.md and existing source code returned by the pythia src tool.
Boundary markers: The instructions do not define explicit delimiters (e.g., XML tags or triple quotes) to separate ingested code from the agent's instructions, though it does mandate use of a linter (pythia lint).
Capability inventory: The skill utilizes a specialized toolset (pythia commands) for metadata discovery and relies on a separate controlled skill (pythia-apply) for database execution.
Sanitization: The skill provides clear guidance on sanitizing string literals using pythia unistr and preventing SQL injection via bind variables.
[SAFE]: The skill proactively addresses common PL/SQL security vulnerabilities by providing a patterns reference that demonstrates how to avoid SQL injection through EXECUTE IMMEDIATE ... USING and ensures robust error handling by discouraging empty WHEN OTHERS exception blocks.

pythia-write