agent-fact-system
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| AFS_HOME | Yes | Absolute path to the AFS data directory. | |
| SILICONFLOW_API_KEY | No | API key for SiliconFlow remote embeddings (optional, required only for vector/hybrid retrieval). |
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": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| fact_getB | Read one canonical AFS fact or practice record. |
| fact_searchC | Search AFS through the existing canonical-adjudicated query service. |
| fact_proposeC | Propose a candidate fact or practice through the canonical service. |
| timelineC | Read canonical Timeline events with optional filters. |
| healthB | Run the existing offline AFS doctor report. |
| statsA | Read the existing local index status and canonical watermark statistics. |
| document_getC | Read a document when the optional document capability is installed. |
| document_listB | List documents when the optional document capability is installed. |
| document_searchC | Search documents when the optional document capability is installed. |
| document_putC | Put a document when the optional document capability is installed. |
| document_resolveC | Resolve a partial document slug when the capability is installed. |
| document_tagsC | List tags for a document when the capability is installed. |
| document_linksB | List outgoing document links when the capability is installed. |
| document_backlinksB | List incoming document links when the capability is installed. |
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 14 tools
Each tool targets a distinct resource/action: fact_get, fact_search, and fact_propose cover fact reading and proposing; timeline, health, and stats are separate system views; the eight document_* tools each perform a unique operation on documents. There is only a very minor overlap between health and stats, but their descriptions clearly differentiate them.
The fact_* and document_* tools use a prefix-plus-verb pattern (fact_get, document_put), but timeline, health, stats are bare nouns, and document_tags, document_links, and document_backlinks are noun-noun constructions. The mixed conventions are readable but not consistent enough for a high score.
Fourteen tools is within the well-scoped range for a system that manages facts and optional document capabilities. Each tool has a clear purpose, and the count is reasonable for the domain.
The fact lifecycle covers get, search, and propose, which is appropriate for a canonical/adjudicated system. Document tools cover get, list, search, put, resolve, tags, links, and backlinks, but there is no explicit delete operation for documents or facts, which is a minor gap.