19 tools from the Phoenix MCP Server, categorised by risk level.
View the Phoenix policy →get-dataset-examples Fetch examples from a specific dataset 2/5 get-dataset-experiments List experiments run on a dataset 2/5 get-experiment-by-id Get full experiment metadata and results 2/5 get-latest-prompt Get the most recent version of a prompt 2/5 get-prompt-by-identifier Retrieve a prompt by its identifier 2/5 get-prompt-version Get a specific version of a prompt 2/5 get-prompt-version-by-tag Retrieve a prompt version by tag name 2/5 get-span-annotations Get annotations for a list of span IDs 2/5 get-spans Retrieve spans from a project with filtering 2/5 list-datasets List all datasets in the Phoenix instance 2/5 list-experiments-for-dataset List all experiments for a dataset ID 2/5 list-projects List all projects with optional pagination 2/5 list-prompt-version-tags List all tags applied to prompt versions 2/5 list-prompt-versions List all versions of a given prompt 2/5 list-prompts List all prompt templates in the project 2/5 phoenix-support Search Phoenix documentation via RunLLM 2/5 The Phoenix MCP server exposes 19 tools across 2 categories: Read, Write.
Use Intercept, the open-source MCP proxy. Write YAML rules for each tool — rate limits, argument validation, or deny rules — then run Intercept in front of the Phoenix server.
Phoenix tools are categorised as Read (16), Write (3). Each category has a recommended default policy.
Open source. One binary. Zero dependencies.
npx -y @policylayer/intercept