F
licenseNot graded
qualityC
maintenanceA local-first MCP server that manages developer memory for coding agents, enabling shared project context, permissions, and audit trails across different agents.
1
-
MemoryGuard
by irisxc4
Related Servers
Alternatives to MemoryGuard
No user-submitted related servers found.
Related Servers
- FlicenseNot gradedqualityDmaintenanceShared memory and orchestration for coding agents, enabling persistent knowledge, multi-agent coordination, and a canonical workflow across MCP-compatible AI clients.10110-
- AlicenseNot gradedqualityAmaintenanceEnables AI coding agents to share a local-first, versioned memory of decisions, conventions, tasks, conflicts, and handoffs over MCP and REST.2MIT
- AlicenseAqualityAmaintenanceMCP-native, local-first memory for coding agents that turns real sessions into reusable decisions, gotchas, and domain knowledge.176MIT
- AlicenseAqualityCmaintenanceA local-first MCP server that provides a shared Markdown-based memory for AI coding agents, enabling cross-agent context persistence via tools like memory_search and memory_capture.101MIT
- AlicenseNot gradedqualityCmaintenanceAn MCP-native, local-first memory server that gives AI agents persistent, structured memory across sessions and tools, enabling them to maintain identity and context without reconfiguration.3MIT
TDQS
C2.6/5.0
Scored across 61 tools
Disambiguation2/5
Most core categories (memory CRUD, codegraph queries, rule merges) have distinct purposes, but several tools are nearly indistinguishable from their descriptions: history_search, history_timeline, history_read, and history_extract_preview are all described as 'V2-native read surface,' and the five knowledge_* tools offer no meaningful differentiation. With 61 tools, this ambiguity forces an agent to do a lot of guesswork.
Naming Consistency3/5
All tools consistently use the memoryguard_ prefix and snake_case, which helps. However, the pattern flips between domain_action (memory_read, rule_undo, codegraph_query) and action_noun (list_sources, scan_summary, extract_memories, accept_candidates), and several names are bare nouns or adjectives (neuron_graph, codegraph_path, codegraph_affected, knowledge_book). The naming is readable but not predictably regular.
Tool Count1/5
61 tools is an extreme count for a single MCP server, far exceeding the 'too many' threshold. Even with strong subdomain organization, an agent must navigate dozens of near-siblings across memory, rule merging, codegraph, enrichments, history, and knowledge, which creates a severe selection burden.
Completeness4/5
The governed memory, rule, enrichment, codegraph, and diagnostics lifecycles are extensively covered, including undo operations, merge previews, and pipeline status checks. Minor gaps exist, such as binding unbind being CLI/GUI-only and external MCP descriptors lacking update/delete tools, but core workflows are not dead-ended.
Maintenance
ActivityActive
ResponsivenessNo issues