context-ledger
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
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
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| get_file_contextC | Get active rules for files, including rules attached to their parent directories. |
| search_memoryC | Search by 1–3 exact tags, a broad free-text phrase, or both. |
| record_memoryC | Record durable knowledge, optionally scoped to project-relative files or directories. |
| supersede_memoryB | Mark obsolete memory superseded, optionally linking its active replacement. |
| dispute_memoryB | Mark a record disputed when evidence conflicts but no replacement is established. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| context_ledger_instructions | Diagnostic copy of the server-wide instructions sent during MCP initialization. |
TDQS
Scored across 5 tools
The four memory tools (search, record, supersede, dispute) each map to a clearly distinct lifecycle action, and descriptions reinforce the boundaries. The only mild overlap is between search_memory and get_file_context, both being retrieval tools, but the file-scoped/parent-directory semantics of get_file_context make them distinguishable.
All names follow a clean snake_case verb_noun pattern (search_memory, record_memory, supersede_memory, dispute_memory, get_file_context) with no mixing of conventions or vague verbs. The pattern is fully predictable.
Five tools is well-scoped for a memory ledger: one create, two retrieval paths, and two state-transition operations. Each tool earns its place without redundancy or obvious bloat.
The surface covers the core ledger lifecycle: recording, searching, retiring (supersede), and flagging (dispute), plus file-scoped retrieval. Minor gaps exist, such as retrieving a single record by id or listing all memories, but these are workarounds via search rather than hard dead ends.