Skip to main content
Glama

Ace Memory

by io.github.rudedoggg
OverviewDiscussionsAdmin

Server Details

Persistent memory for AI agents. Semantic search, memory graph, W3C DID identity.

Glama couldn't complete the latest health check. If this server requires authentication, missing or expired test credentials may be the cause. A test profile lets Glama authenticate for health checks and discover tools; it is separate from your personal connections.

If you are the author, claim ownership, then add or update a test profile under Admin → Test Profile.

Status
Unhealthy
OAuth
Works in Glama
Last Tested
Transport
Streamable HTTP
URL

TDQS

B3.1/5.0

Scored across 82 tools

Disambiguation2/5

The tool set has significant overlap and ambiguity, particularly in memory and context management. For example, 'agent_memory_store', 'remember', and 'observe' all handle memory storage with unclear distinctions, while 'assemble_context' and 'session_start' both assemble context but with different approaches. This overlap makes it difficult for an agent to reliably choose the correct tool without deep domain knowledge.

Naming Consistency4/5

Most tools follow a consistent verb_noun or noun_verb pattern (e.g., 'agent_create', 'memory_export', 'privacy_check_access'), with clear prefixes like 'agent_', 'memory_', 'skill_' for grouping. However, there are minor deviations such as 'link', 'observe', 'recall', and 'whoami' that break this pattern, slightly reducing overall consistency.

Tool Count2/5

With 82 tools, the count is excessive for a memory management server, leading to bloat and potential confusion. Many tools could be consolidated (e.g., multiple memory storage and context assembly tools) or omitted without losing functionality. This large number overwhelms the core purpose and suggests poor scoping.

Completeness5/5

The tool set provides comprehensive coverage for agent memory management, including creation, updating, deletion, benchmarking, dreaming, privacy, skills, and context assembly. It supports full CRUD operations across agents, memories, goals, and other entities, with no apparent gaps in the domain's lifecycle or workflows.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

  1. 1 tool update
    • Changedagent_load1 field changed
      • changedInput schema / properties / query / description
        Previous value: -"Agent name to look up (e.g., \"Corbot\", \"code-assistant\")"New value: +"Agent name to look up (e.g., \"Corbot\", \"Data\", \"code-assistant\")"
  2. 1 tool update
    • Addedagent_load
  3. 1 tool update
    • Changedagent_memory_query5 fields changed
      • addedInput schema / properties / asOf
        Added value: +{
+  "description": "ISO 8601 timestamp — return memories valid at this point in time. Use for temporal reasoning (\"what did I know on date X?\").",
+  "type": "string"
+}
      • addedInput schema / properties / expandRelationships
        Added value: +{
+  "default": false,
+  "description": "Expand initial results via the memory relationship graph. Surfaces connected memories (CAUSED_BY, ELABORATES, RELATED_TO, etc.). Use when context-rich recall matters more than precision.",
+  "type": "boolean"
+}
      • addedInput schema / properties / includeSuperseded
        Added value: +{
+  "default": false,
+  "description": "Include memories that have been superseded by newer ones. Default: false (cleaner results, better recall on knowledge-update questions).",
+  "type": "boolean"
+}
      • addedInput schema / properties / minRelationshipWeight
        Added value: +{
+  "default": 0.3,
+  "description": "Minimum edge weight to include during graph traversal. Default: 0.3.",
+  "maximum": 1,
+  "minimum": 0,
+  "type": "number"
+}
      • addedInput schema / properties / relationshipDepth
        Added value: +{
+  "default": 1,
+  "description": "Graph traversal depth when expandRelationships is true. 1=direct neighbors, 2=neighbors-of-neighbors. Default: 1.",
+  "enum": [
+    1,
+    2
+  ],
+  "type": "number"
+}
  4. 4 tool updates
    • Addedmemory_consolidate
    • Addedmemory_curate
    • Addedmemory_forget
    • Addedmemory_promote
  5. 82 tool updates
    • First observedagent_benchmark_history
    • First observedagent_benchmark_run
    • First observedagent_benchmark_trend
    • First observedagent_context_clear
    • First observedagent_context_get
    • First observedagent_create
    • First observedagent_delete
    • First observedagent_dream
    • First observedagent_dream_config
    • First observedagent_dream_history
    • First observedagent_facts_contradictions
    • First observedagent_facts_create
    • First observedagent_facts_list
    • First observedagent_facts_search
    • First observedagent_facts_update
    • First observedagent_get
    • First observedagent_handoff_initiate
    • First observedagent_handoff_latest
    • First observedagent_handoff_resume
    • First observedagent_link_verify
    • First observedagent_list
    • First observedagent_memory_expand
    • First observedagent_memory_query
    • First observedagent_memory_store
    • First observedagent_transfer
    • First observedagent_update
    • First observedassemble_context
    • First observedauthorize
    • First observedconstitution_check_conflicts
    • First observedconstitution_get
    • First observedconstitution_get_tier
    • First observedconstitution_list_proposals
    • First observedconstitution_propose_change
    • First observedconstitution_validate_action
    • First observedcontext_budget_apply_preset
    • First observedcontext_budget_get
    • First observedcontext_budget_set
    • First observedgoal_complete
    • First observedgoal_create
    • First observedgoal_find_similar
    • First observedgoal_get
    • First observedgoal_get_success_rate
    • First observedgoal_list
    • First observedgoal_update_progress
    • First observedlink
    • First observedmemory_adapters
    • First observedmemory_audit
    • First observedmemory_clean
    • First observedmemory_export
    • First observedmemory_import
    • First observedmemory_ingest
    • First observedmemory_ingest_batch
    • First observedmemory_source_approve
    • First observedmemory_source_list
    • First observedmemory_source_register
    • First observedmemory_source_revoke
    • First observedmemory_sync
    • First observedobserve
    • First observedprivacy_check_access
    • First observedprivacy_create_presentation
    • First observedprivacy_disclosure_log
    • First observedprivacy_grant_consent
    • First observedprivacy_list_grants
    • First observedprivacy_revoke_consent
    • First observedrecall
    • First observedregister
    • First observedremember
    • First observedsession_start
    • First observedskill_generate_lessons
    • First observedskill_get_effectiveness
    • First observedskill_get_ranking
    • First observedskill_list
    • First observedskill_load
    • First observedskill_loader_explain
    • First observedskill_loader_recommend
    • First observedskill_loader_resolve
    • First observedskill_loader_resolve_multi
    • First observedskill_recommend
    • First observedskill_record_usage
    • First observedskill_unload
    • First observedtrust_status
    • First observedwhoami

Related MCP Connectors

Related MCP Servers

  • MemoryGate

    PStryder
    A
    license
    Not graded
    quality
    B
    maintenance
    Provides durable memory for AI agents with structured storage, semantic search, OAuth authentication, and lifecycle controls.
    11
    Apache 2.0
Try in Browser
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources