Pmll Memory Mcp
A persistent memory layer for Claude that combines short-term key-value storage with Q-promise deduplication to avoid redundant operations. Built as a companion to the PPM Python package manager, it integrates Context+ semantic memory graphs to maintain conversation state and project context across sessions. Reach for this when you need Claude to remember decisions, cache expensive computations, or maintain awareness of prior work without re-processing. The MCP exposes stdio transport and plugs into Claude Desktop or any MCP-compatible client. Useful for long-running agent tasks where context continuity matters more than ephemeral chat.
- GitHub stars
- 14
- tools
- 15
- transport
- STDIO
- last updated
- Mar 8, 2026
Tools
15 toolsPublic tool metadata for what this MCP can expose to an agent.
initSet up PMLL silo and Q-promise state for a session. Call once at task start.
Set up PMLL silo and Q-promise state for a session. Call once at task start.
No parameter schema in public metadata yet.
peekNon-destructive cache lookup + Q-promise check. Call before every expensive MCP tool invocation.
Non-destructive cache lookup + Q-promise check. Call before every expensive MCP tool invocation.
No parameter schema in public metadata yet.
setStore a key-value pair in the session silo. Call after a cache miss.
Store a key-value pair in the session silo. Call after a cache miss.
No parameter schema in public metadata yet.
resolveCheck/resolve a Q-promise continuation
Check/resolve a Q-promise continuation
No parameter schema in public metadata yet.
flushClear all silo slots at task completion
Clear all silo slots at task completion
No parameter schema in public metadata yet.
graphqlExecute GraphQL queries/mutations against the memory store
Execute GraphQL queries/mutations against the memory store
No parameter schema in public metadata yet.
upsert_memory_nodeCreate or update a memory node with auto-generated stable hashing embeddings
Create or update a memory node with auto-generated stable hashing embeddings
No parameter schema in public metadata yet.
create_relationCreate typed edges (relates_to, depends_on, implements, references, similar_to, contains)
Create typed edges (relates_to, depends_on, implements, references, similar_to, contains)
No parameter schema in public metadata yet.
search_memory_graphSemantic search with graph traversal — direct matches + neighbor walk
Semantic search with graph traversal — direct matches + neighbor walk
No parameter schema in public metadata yet.
prune_stale_linksRemove decayed edges (e^(-λt) below threshold) and orphan nodes with low access
Remove decayed edges (e^(-λt) below threshold) and orphan nodes with low access
No parameter schema in public metadata yet.
add_interlinked_contextBulk-add nodes with auto-similarity linking (cosine ≥ 0.72 creates edges)
Bulk-add nodes with auto-similarity linking (cosine ≥ 0.72 creates edges)
No parameter schema in public metadata yet.
retrieve_with_traversalWalk outward from a node — returns reachable neighbors scored by decay & depth
Walk outward from a node — returns reachable neighbors scored by decay & depth
No parameter schema in public metadata yet.
resolve_contextUnified context lookup: short-term KV → long-term graph → miss
Unified context lookup: short-term KV → long-term graph → miss
No parameter schema in public metadata yet.
promote_to_long_termPromote a short-term KV entry to the long-term memory graph
Promote a short-term KV entry to the long-term memory graph
No parameter schema in public metadata yet.
memory_statusUnified view of short-term KV and long-term graph memory status
Unified view of short-term KV and long-term graph memory status
No parameter schema in public metadata yet.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageReadme
PPM — Python Package Manager
pypm packaging CLI with shared PMLL memory + Q-promise. Sibling: drQedwards/pmll.
Live Stellar pmll-anchor IDs: docs/STELLAR.md · skills: SKILL.md · promises: Q_promise_lib/README.md.
Details
- Repository
- github.com/drqedwards/ppm
- Category
- AI & LLM Tools
- Package
pmll-memory-mcp- Transport
- STDIO
- Registry
- MCP registry, active
Related AI & LLM Tools MCP servers
View all
Agent Emailusejunior/email-agent-mcpLocal email connectivity for AI agents — read, draft, send, and organize Outlook mail via MCP7stars
A2dbagentic-eng/a2dbSafe, read-only database access for AI agents. 5 databases, batch queries, SQLGlot.2stars
Elfsight AI Chatbotelfsight/ai-chatbot-mcpRead conversations, leads, knowledge base and analytics of your Elfsight AI Chatbot widgets.2stars
Agent Readymlava/agent-ready-mcpScan any URL for AI agent readability — Vercel Spec, llmstxt.org, and agent-protocol manifests.2stars
Agent Audit Mcprumblingb/agent-audit-mcpImmutable audit trail for agent-to-agent interactions2stars