Daemon-MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| DAEM0NMCP_LOG_LEVEL | No | Logging level | INFO |
| DAEM0NMCP_PROJECT_ROOT | No | Project root path | . |
| DAEM0NMCP_STORAGE_PATH | No | Override storage location (auto-detected by default) | |
| DAEM0NMCP_HYBRID_VECTOR_WEIGHT | No | Configurable hybrid weight for search (0.0-1.0) | |
| DAEM0NMCP_SEARCH_DIVERSITY_MAX_PER_FILE | No | Limits same-file results for result diversity |
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": true
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| extensions | {
"io.modelcontextprotocol/ui": {}
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| simulate_decisionA | Temporal Scrying -- replay a past decision with current knowledge. Reconstructs the context that existed at the decision's moment of inscription, compares it with the daemon's present understanding, and reveals what is now known that was not known then. Args: decision_id: Memory ID of the decision to scry project_path: Project root |
| evolve_ruleA | Rule Entropy Analysis -- examine rules for signs of decay and drift. Cross-references rule triggers against the code index and outcome history to produce staleness scores and concrete evolution suggestions. The daemon proposes changes but never inscribes them without consent. Args: rule_id: Specific rule ID to analyze (omit for batch analysis of all rules) project_path: Project root |
| debate_internalA | Adversarial Council -- convene an internal debate grounded in memory evidence. Runs a structured advocate/challenger/judge debate where each argument is supported only by recalled memory evidence. No external reasoning is invoked. Convergence detection halts deliberation when positions stabilize. The synthesis is inscribed as a consensus memory for future retrieval. Args: topic: The subject of deliberation advocate_position: The position the advocate will defend challenger_position: The position the challenger will defend project_path: Project root |
| communeC | Session start & status operations. Actions: briefing, active_context, triggers, health, covenant, updates |
| consultD | Pre-action intelligence gathering. Actions: preflight, recall, recall_file, recall_entity, recall_hierarchical, search, check_rules, compress |
| inscribeC | Memory writing & linking operations. Actions: remember, remember_batch, link, unlink, pin, activate, deactivate, clear_active, ingest |
| reflectD | Outcomes & verification operations. Actions: outcome, verify, execute |
| understandC | Code comprehension operations. Actions: index, find, impact, todos, refactor |
| governC | Rules & triggers management. Actions: add_rule, update_rule, list_rules, add_trigger, list_triggers, remove_trigger |
| exploreC | Graph & discovery operations. Actions: related, chain, graph, stats, communities, community_detail, rebuild_communities, entities, backfill_entities, evolution, versions, at_time |
| maintainC | Housekeeping & federation operations. Actions: prune, archive, cleanup, compact, rebuild_index, export, import_data, link_project, unlink_project, list_projects, consolidate, purge_dream_spam |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| Test UI | Test UI to validate MCP Apps infrastructure |
TDQS
Scored across 11 tools
Each tool targets a distinct operational domain (memory, rules, code, decisions, housekeeping), with only minor overlap between explore and consult in retrieval contexts. Descriptions are clear enough to differentiate them.
Tool names mix single verbs (reflect, explore, commune) with compound forms (simulate_decision, evolve_rule, debate_internal). This inconsistent pattern makes it harder to predict tool function from name alone.
11 tools is well within the ideal 3-15 range, and each tool bundles a coherent set of sub-operations that justifies its presence. The count feels appropriate for the server's broad scope.
The surface covers memory, rules, code comprehension, decision analysis, session management, and housekeeping. Minor gaps include lack of explicit rule deletion and some overlapping retrieval methods, but these do not significantly hinder agent workflows.