EvidentTrail MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| EVIDENTTRAIL_HOST | No | HTTP hook server bind address. Warning: non-localhost values expose an unauthenticated /hook endpoint. | 127.0.0.1 |
| EVIDENTTRAIL_PORT | No | HTTP hook server port | 3100 |
| EVIDENTTRAIL_API_KEY | Yes | Long-lived API key (et_live_...) | |
| EVIDENTTRAIL_API_URL | Yes | Base URL of the EvidentTrail API, e.g. https://evidenttrail-api.fly.dev | |
| EVIDENTTRAIL_MODEL_ID | No | Optional model ID, e.g. claude-sonnet-4-6 | |
| EVIDENTTRAIL_AGENT_NAME | No | Display name for the agent | claude-code |
| EVIDENTTRAIL_AGENT_VERSION | No | Optional version string | |
| EVIDENTTRAIL_REPOSITORY_ID | Yes | UUID of the repository being worked on |
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 | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| et_log_actionA | Log an agent action as compliance evidence in EvidentTrail. Call this whenever you perform a significant action (file edit, API call, decision). |
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 1 tool
With a single tool there is no possibility of misselection; the tool's purpose (logging an agent action as compliance evidence) is unambiguous and clearly stated. No overlap risk exists in the current surface.
The name et_log_action follows a clear, predictable snake_case verb_noun pattern with a consistent server prefix. Though there is only one tool, the convention is well-defined and would scale cleanly.
A single tool is extremely thin for a compliance-evidence server, which typically implies writing, querying, and verifying records. One write-only operation cannot cover the apparent scope of the domain.
The surface only supports appending evidence; there is no way to retrieve, search, list, verify, or export logged actions. These gaps would leave an agent unable to confirm or audit what was recorded, a dead end for a compliance workflow.