The Agent Skills Directory

[PROMPT_INJECTION]: The skill establishes a surface for indirect prompt injection by instructing the agent to read and modify Python backend code in the phoenix repository.
Ingestion points: Files in src/phoenix/server/, src/phoenix/db/, and tests/unit/server/.
Boundary markers: The provided templates do not include delimiters or specific instructions to ignore embedded prompts within the processed code.
Capability inventory: The agent is empowered to modify codebase files and execute shell commands such as make, uv, and pytest.
Sanitization: No specific sanitization or validation of the codebase content is described.
[COMMAND_EXECUTION]: The skill includes instructions to execute various local development, build, and test commands.
Evidence: Commands such as make dev-backend, uv run pytest, make graphql, make format, make typecheck-python, and uv run alembic revision. These represent standard development practices using established tools.

phoenix-server