Gmem
This gives AI agents persistent memory for Solana and EVM projects across sessions. Seven MCP tools backed by SQLite: recall queries project state via BM25 search, write stores typed entities like Program, Account, PDA, Decision, Finding. The diff tool accepts git refs (HEAD~3, branch names) not just timestamps. Auto-ingest tools parse Anchor.toml and Hardhat configs to populate program IDs, IDL hashes, deployment addresses, and chain IDs without manual entry. Solana CLI context capture auto-attributes decisions to your active wallet and cluster. Reach for it when your agent keeps forgetting the program ID it deployed yesterday or why you rejected a specific integration last week.
- GitHub stars
- 1
- tools
- 9
- transport
- STDIO
- env var
- 1
- last updated
- May 17, 2026
Tools
9 toolsPublic tool metadata for what this MCP can expose to an agent.
gmem.recallRetrieve memory entries relevant to a natural-language query, ranked by BM25 + recency
Retrieve memory entries relevant to a natural-language query, ranked by BM25 + recency
No parameter schema in public metadata yet.
gmem.writePersist a typed memory entry (Program / Account / Instruction / Decision / Finding / Integration); append-only
Persist a typed memory entry (Program / Account / Instruction / Decision / Finding / Integration); append-only
No parameter schema in public metadata yet.
gmem.diffShow how memory state changed between two points in time — accepts ISO timestamps OR git commit refs (HEAD, HEAD~3, branch names, full or short SHAs)
Show how memory state changed between two points in time — accepts ISO timestamps OR git commit refs (HEAD, HEAD~3, branch names, full or short SHAs)
No parameter schema in public metadata yet.
gmem.list_decisionsList all Decision entries for the active project, newest first
List all Decision entries for the active project, newest first
No parameter schema in public metadata yet.
gmem.ingest_anchorAuto-ingest an Anchor workspace: parse Anchor.toml, capture IDL sha256s from target/idl/, record the current git HEAD as source commit, write one Program per (program, cluster) pair
Auto-ingest an Anchor workspace: parse Anchor.toml, capture IDL sha256s from target/idl/, record the current git HEAD as source commit, write one Program per (program, cluster) pair
No parameter schema in public metadata yet.
gmem.solana_contextRead the active Solana CLI config (~/.config/solana/cli/config.yml), return the configured cluster + RPC URL + active-keypair pubkey. Never returns the secret key.
Read the active Solana CLI config (~/.config/solana/cli/config.yml), return the configured cluster + RPC URL + active-keypair pubkey. Never returns the secret key.
No parameter schema in public metadata yet.
gmem.ingest_hardhatAuto-ingest a Hardhat / EVM workspace: parses hardhat.config, reads deployment artifacts, classifies the network into a canonical chain, records a reorder-invariant ABI SHA-256 and git HEAD as sourceCommit. Writes one Contract entity per (chain, address) pair.
Auto-ingest a Hardhat / EVM workspace: parses hardhat.config, reads deployment artifacts, classifies the network into a canonical chain, records a reorder-invariant ABI SHA-256 and git HEAD as sourceCommit. Writes one Contract entity per (chain, address) pair.
No parameter schema in public metadata yet.
gmem.ingest_agent_activityPull Solana activity for an autonomous-agent wallet via Helius RPC, aggregate it, score it with the transparent v1 heuristic, and write one Agent entity per (cluster, pubkey).
Pull Solana activity for an autonomous-agent wallet via Helius RPC, aggregate it, score it with the transparent v1 heuristic, and write one Agent entity per (cluster, pubkey).
No parameter schema in public metadata yet.
gmem.score_agentReturn the latest stored Agent trust score with the heuristic breakdown showing which factors moved it. Read-only: no RPC call, no write.
Return the latest stored Agent trust score with the heuristic breakdown showing which factors moved it. Read-only: no RPC call, no write.
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 languageConfiguration
GMEM_DBOverride the auto-resolved memory database path. Defaults to ~/.gmem/<projectHash>/memory.db, with one db file per Solana project (project root resolved from Anchor.toml or workspace Cargo.toml).
Readme
gmem
Persistent project memory for Solana AI agents.
A Model Context Protocol (MCP) server that gives AI coding agents — Claude Code, Cursor, or anything that speaks MCP — durable, Solana-aware memory of a project across sessions: program IDs, IDLs, PDA seeds, deployment state, architectural decisions, audit findings. So agents stop forgetting what they built yesterday.
Details
- Repository
- github.com/yksanjo/gmem
- Categories
- AI & LLM Tools, Developer Tools, Finance & Commerce
- Package
@yksanjo/gmem- Transport
- STDIO
- Registry
- MCP registry, active
Related AI & LLM Tools MCP servers
View all
Deeprepoabhishek2432001/deeprepoProductivity-boosting RAG engine for codebases with multi-provider AI support and semantic search.0stars
X402engine Mcpagentc22/x402engine-mcp50+ pay-per-call APIs for AI agents — images, LLM, code, crypto, travel & more via x4020stars
Agentic Terminal Directoryobserver-protocol/at-directoryVerified merchants accepting agentic payments on Lightning/L402/BOLT12/USDT — search, verify, pay.0stars
Computebackautonomad1/computeback-mcpAgent Rewards Marketplace: earn $NOMD on B2B work, spend on agent capabilities.0stars
Not Human Searchai.nothumansearch/searchSearch the agentic web. 4,100+ sites, 11 tools incl. check_url + verify_mcp for probe-before-use.0stars
El Buen Agente Mcpapasztetnik/el-buen-agente-mcpLa guía 'El Buen Agente' como 18 tools para evaluar, mejorar y construir agentes LLM. En español.0stars