Skip to main content
Glama
madbradsmith

DUMEdb

by madbradsmith
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Local

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
PYTHONUNBUFFEREDNoSet to "1" in the MCP server environment configuration to ensure unbuffered Python output over stdio.1

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

CapabilityDetails
tools
{
  "listChanged": false
}
prompts
{
  "listChanged": false
}
resources
{
  "subscribe": false,
  "listChanged": false
}

Tools

Functions exposed to the LLM to take actions

NameDescription
dume_read_stateB
Direct zero-copy LMDB read of active state registers for a given entity_id.
Decodes 32-byte records via SchemaDictionary into a human-readable dictionary.
dume_queryC

High-speed associative search across active state registers and extension payloads in DUMEdb. Scans hardware-aligned LMDB state without touching raw filesystem documents or burning token context. Matches opcodes, enum values, structured register names, and text payloads.

dume_commit_stateB
Atomically commits a discrete state register mutation to DUMEdb LMDB storage.
Resolves opcode_name from SchemaDictionary (or auto-registers novel schema)
and packs value into a hardware-aligned 32-byte record.
dume_list_schemasA
Returns active DUMEdb schema dictionary specifications, including both pre-registered
and auto-discovered [AUTO] opcodes.
dume_flush_scratchpadB
Resets the conversational scratchpad buffer for the specified entity_id
while keeping persistent DUMEdb LMDB registers intact.
dume_read_adrB
Returns the content of the target Architectural Decision Record (ADR) markdown file
from ~/.dume/assets/adrs/.
dume_get_error_detailsB
Returns the sanitized stack trace and diagnostic context for an active error fingerprint card
from ~/.dume/assets/errors/.
dume_get_incident_rcaB
Returns the root cause analysis and resolution runbook for a logged incident card
from ~/.dume/assets/incidents/.
dume_list_module_contextB

Returns all known ADRs, active error fingerprints, and unresolved incidents linked to the specified module or system area.

dume_ingest_enterpriseA
Ingests an enterprise JSON payload (PR design proposal, APM exception stack trace, or incident postmortem ticket).
Scrubs secrets/PII, generates a Markdown card in ~/.dume/assets/, and commits active state pointers to DUMEdb.
cabinet_wakeA
CRITICAL FIRST STEP: Executes the Wake Handshake.
MANDATORY: You MUST call this tool at the very beginning of every new session, or whenever
the user greets you, asks for status, says 'catch me up', 'wake', 'start', 'briefing',
or uses the phrase 'box is the game'.
DO NOT search the filesystem, inspect files, or take any other actions until this tool has been executed.
Validates codeword, ingests compass (status_block.md + last_session.md) in < 250 tokens,
checks DUMEdb state, and returns a tight briefing package with zero bloat.
cabinet_list_workflowsB
Returns high-density, low-token list of active heuristics in The Cabinet.
Categories: meta, shape, speed, self_monitoring, posture, protocol, custom.
Use this to pull operational tools without reading thousands of prompt tokens.
cabinet_record_workflowB
Records a new workflow heuristic in The Cabinet when a repeatable pattern has earned its place (Rule WF-M3).
Categories: meta, shape, speed, self_monitoring, posture, protocol, custom.
Automatically assigns next ID, persists to structured ledger, and syncs workflow_notes.md.
cabinet_amend_workflowB

Amends or retires an existing workflow heuristic in The Cabinet. Pass retire=True to mark a heuristic as retired when superseded or no longer firing.

cabinet_refreshA
Executes the Refresh Protocol (RP) when context hits ~70%.
Writes last_session.md, updates status_block.md, commits state, and confirms
KV-cache is ready for a clean restart.
Optionally records new or retired workflow heuristics discovered during the session.

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

Deploy Server
Context All Your Agents Can ShareAgents share what they learn, hand off work, and hold it when you switch models. Start with 5 GB free.fulcradynamics.comAd

TDQS

A3.5/5.0

Scored across 15 tools

Disambiguation4/5

Most tools target clearly distinct resources or lifecycle phases: state registers, ADR/error/incident cards, workflow heuristics, and session wake/refresh. Slight ambiguity exists between dume_read_state and dume_query, and between cabinet_wake and cabinet_refresh, but descriptions make their roles reasonably clear.

Naming Consistency4/5

Names are consistently namespaced with dume_ or cabinet_ and mostly follow a verb_noun pattern such as list_module_context, commit_state, record_workflow, and get_error_details. A few single-verb names like dume_query, cabinet_wake, and cabinet_refresh are minor deviations but still readable.

Tool Count5/5

With 15 tools, the set is at the upper end of the ideal range but each tool appears to serve a distinct purpose across state management, retrieval, ingestion, session lifecycle, and workflow heuristics. There is no obvious redundancy or bloat.

Completeness4/5

The surface covers the core lifecycle: wake/refresh, state read/query/commit, enterprise ingestion, ADR/error/incident retrieval, schema listing, and workflow management. Minor gaps remain, such as no explicit entity/module listing or direct ADR/error/incident mutation beyond ingestion, but core workflows are workable.

Maintenance

ActivityMaintained
ResponsivenessNo issues

Resources