phab-arc-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| CONDUIT_TOKEN | No | Optional alias for PHAB_API_TOKEN. | |
| PHAB_API_TOKEN | No | Primary API token for Phabricator Conduit. | |
| PHAB_CONDUIT_URI | No | Base URL of your Phabricator instance. | https://phab.instahyre.com/ |
| PHAB_CONDUIT_TOKEN | No | Optional alias for PHAB_API_TOKEN. | |
| PHAB_ARC_TIMEOUT_MS | No | Request timeout in milliseconds. | 30000 |
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 | {} |
| prompts | {} |
| resources | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| inline-comments-phabA | Creates draft inline comments on a Differential revision from review findings JSON (does not publish). |
| review-phabA | Fetches the exact review prompt and revision context needed to review a Differential without MCP sampling. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| review-phab | Structured Differential review prompt that uses provided revision context and returns JSON findings. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 2 tools
The two tools have clearly distinct purposes: one fetches review context, the other creates draft inline comments. There is no overlap in functionality, so an agent can easily select the right tool.
Tool names do not follow a consistent pattern. 'review-phab' is verb-noun, while 'inline-comments-phab' is noun-noun. The shared '-phab' suffix is a minor unifier but the lack of consistent verb usage makes naming unpredictable.
With only two tools, the server feels thin but not unreasonable for a narrowly focused review workflow. The count is at the borderline where it could benefit from additional operations, but it is not excessively sparse.
The tool surface has significant gaps. It creates draft comments but explicitly does not publish them, and there is no way to submit or finalize a review. This creates a dead end for the agent workflow, as the created comments cannot be acted upon fully.