@git-fabric/chat
OfficialServer Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| QDRANT_URL | Yes | Qdrant Cloud cluster URL | |
| GITHUB_TOKEN | Yes | GitHub PAT for state repo read/write | |
| OPENAI_API_KEY | Yes | OpenAI API key for text-embedding-3-small | |
| QDRANT_API_KEY | Yes | Qdrant Cloud API key | |
| ANTHROPIC_API_KEY | Yes | Anthropic API key for Claude completions | |
| GITHUB_STATE_REPO | No | State repo (default: ry-ops/git-steer-state) | ry-ops/git-steer-state |
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 |
|---|---|
| chat_session_createA | Create a new chat session with Claude. Optionally set a system prompt, project tag, model, and title. Returns the sessionId to use in subsequent calls. |
| chat_session_listA | List recent chat sessions. Filter by project, state, and limit. Sessions are sorted by most recently updated first. |
| chat_session_getA | Get full session details including message history. Use this to inspect or resume a prior conversation. |
| chat_session_archiveA | Archive a session. Archived sessions are hidden from the default list but remain searchable and resumable. |
| chat_session_deleteA | Permanently delete a session and all its messages. This also removes vectors from Qdrant. Irreversible. |
| chat_message_sendA | Send a message in an existing session and get a Claude response. Reconstructs full conversation history for the API call. Stores both user message and assistant response. Returns the assistant reply with token usage. |
| chat_message_listB | List messages in a session with pagination. Returns messages in chronological order. |
| chat_searchA | Semantic search over all stored conversation content using vector similarity. Finds messages relevant to the query even if exact words don't match. Optionally scope to a project or specific session. |
| chat_context_injectA | Inject external context into a session before the next message send. Use this to pipe in Aiana memory recall, documentation snippets, or runtime state. The injected content is stored as a message and included in the next completion call. |
| chat_statusA | Return aggregate stats: total sessions, total messages, and tokens consumed today. Useful for quota monitoring and observability. |
| chat_healthA | Ping Anthropic and Qdrant services. Returns latency for each. Use to verify the app is operational before sending messages. |
| chat_thread_forkA | Fork a session at a specific message to explore an alternative branch of conversation. Creates a new session with all history up to and including the fork point. The original session is unchanged. |
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 12 tools
Each tool has a clearly distinct purpose with no significant overlap. For example, chat_message_send handles sending messages, chat_search performs semantic searches, and chat_session_archive manages archiving, all targeting different actions and resources. The descriptions reinforce these distinctions, making tool selection straightforward for an agent.
All tool names follow a consistent 'chat_' prefix with descriptive verb_noun patterns, such as chat_session_create, chat_message_list, and chat_thread_fork. This uniformity enhances readability and predictability, allowing agents to easily infer functionality from the names without confusion.
With 12 tools, the server is well-scoped for managing chat sessions, messages, and related operations. Each tool serves a specific role, from creation and listing to archiving and forking, providing comprehensive coverage without unnecessary bloat. This count aligns perfectly with the domain's complexity.
The toolset offers complete CRUD and lifecycle coverage for chat sessions and messages, including create, list, get, update (via send), archive, and delete operations. Additional tools like search, health checks, and forking address advanced needs, leaving no obvious gaps for typical agent workflows in this domain.