phylax-mcp
OfficialServer Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| PHYLAX_API_KEY | No | Alternative to PHYLAX_API_TOKEN. Required if PHYLAX_API_TOKEN is not provided. | |
| PHYLAX_MCP_PORT | No | Port for HTTP transport only. | 8765 |
| PHYLAX_API_TOKEN | No | Your Phylax API token. Required if PHYLAX_API_KEY is not provided. | |
| PHYLAX_API_BASE_URL | No | Base URL for the Phylax API. | https://api.phyi.dev |
| PHYLAX_MCP_AUDIENCE | No | Expected 'aud' claim on inbound tokens. | |
| PHYLAX_MCP_TRANSPORT | No | Set to 'http' for a shared HTTP server. | stdio |
| PHYLAX_MCP_AUTH_TOKEN | No | Required when PHYLAX_MCP_TRANSPORT is set to 'http'. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| verify_artifactA | Verify a software artifact before installing, executing or depending on it. Accepts a package URL such as pkg:npm/express@4.18.2 or pkg:pypi/requests@2.32.3. Returns ALLOW, WARN or BLOCK with a risk score and findings. Call this before suggesting an install command for any third party package. |
| verify_artifactsA | Verify many artifacts in a single call. Use when checking a dependency list, a lockfile, or several candidate packages at once. Preferred over calling verify_artifact repeatedly. |
| verify_mcp_serverA | Verify a Model Context Protocol server before connecting to it or enabling its tools. Checks provenance, requested permissions, the tool surface for unsafe operations, and known risk signals. Call before adding any third party MCP server. |
| get_attestationA | Fetch the signed attestations behind an artifact verdict. Use when the user asks for evidence, provenance detail, or wants to verify offline. |
| check_policyA | Evaluate an artifact against an organization policy and return the decision. Use when the user asks whether something is allowed by their rules, rather than whether it is generally safe. |
| search_artifactsA | Search Phylax for artifacts by name. Use to resolve a vague package name to a concrete package URL before verifying it. |
| phylax_statusA | Check that the Phylax API is reachable, the token works, and report the current plan and remaining quota. Use when other tools fail, to tell an outage apart from a plan or credential problem. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 7 tools
Each tool targets a distinct aspect of the verification workflow: single vs batch verification, MCP servers, attestations, policy checks, search, and status. Even verify_artifact and verify_artifacts are clearly delineated by cardinality with explicit guidance on when to use each.
The majority use a clear verb_noun pattern (verify_*, get_*, check_*, search_*). The singular phylax_status breaks the pattern by using a noun_verb structure, though it is still readable and not confusing.
At seven tools, the set is well-scoped for a verification service without redundancy or bloat. Each tool addresses a distinct need, and the batch variant is a justified convenience rather than a duplicate.
The core workflow (search, verify, attest, policy-check, status) is well covered. A minor gap is that check_policy handles only artifacts, not MCP servers, so policy evaluation for MCP server permissions is not explicit.