73
/ 100
15 days ago
glama

total-agent-memory

Persistent, local memory for AI coding agents that learns how you work, not just what you said. Supports Claude Code, Codex CLI, Cursor, and any MCP client.

Is this your MCP?

Claim it to get a verified publisher badge, a free copy of our full audit findings, and direct contact for any high-priority issues we find.

Install from

M8ven verifies MCPs across every public registry — install directly from whichever one you prefer.

// key findings
No credential exfiltration, no sensitive file access, no obfuscation
Static analysis found nothing flowing your secrets to unexpected places.
Open source with a license and README
Anyone can audit the code, the license is declared, and the publisher documents what it does.
🔐
You'll be asked for 5 credentials: OPENAI_API_KEY, ANTHROPIC_API_KEY, MEMORY_LLM_API_KEY, COHERE_API_KEY, MEMORY_EMBED_API_KEY
These are read from process.env at runtime. Make sure you trust where they’ll be sent.
// required environment variables
This server reads these from process.env. You'll be asked to provide them before it can run.
🔐 secretOPENAI_API_KEY
🔐 secretANTHROPIC_API_KEY
configCLAUDE_MEMORY_DIR(TAM_MEMORY_DIR canonical, kept as compatibility alias,
configMEMORY_LLM_ENABLEDauto auto\true\false\force — LLM enrichment toggle
configV11_PIPELINE
configV11_SKIP_NLI
configV9_RERANKER_BACKENDReranker backend stays on ce-marco by default. =bge-v2-m3 (or off) switches at runtime.
configOLLAMA_URL
configMEMORY_LLM_PROVIDERexport =openai # or "anthropic"
configMEMORY_LLM_MODELexport =gpt-4o-mini # or "claude-haiku-4-5"
configMEMORY_TRIPLE_PROVIDERexport =openai
configMEMORY_TRIPLE_MODELexport =gpt-4o-mini
🔐 secretMEMORY_LLM_API_KEYexport =sk-...
configMEMORY_ASYNC_ENRICHMENTfalse v10.1 — move quality gate / contradiction / entity dedup / episodic / wiki to a background worker. See [Performance tuning](#performance-tuning)
configMEMORY_QUALITY_GATE_ENABLED
configMEMORY_CONTRADICTION_DETECT_ENABLED
configMEMORY_OUTBOX_ENABLED
configMEMORY_EPISODIC_ENABLED
configREFLECT_DEBOUNCE_SEC
configREFLECT_INTERVAL_SEC
configCLAUDE_TOTAL_MEMORY_SRC
configMCP_TRANSPORT
configMCP_HTTP_HOST
configMCP_HTTP_PORT
configDASHBOARD_PORT37737 HTTP dashboard port
configDASHBOARD_BIND
configTAM_SUPERVISOR_SERVICES
configTAM_MEMORY_DIR( canonical, CLAUDE_MEMORY_DIR kept as compatibility alias,
configTAM_SUPERVISOR_MAX_RESTARTS
configTAM_SUPERVISOR_RESTART_WINDOW
configMEMORY_ACTIVECONTEXT_VAULTexport =~/Documents/project/Projects # default
configMEMORY_NLI_MODEL
configMEMORY_NLI_CALIBRATION_PATH
configV9_CACHE_L1_ENABLED
configV9_CACHE_L2_ENABLED
configV9_CACHE_L1_SIZE
configV9_CACHE_L1_TTL_SEC
configMEMORY_TAG_VOCAB_PATH
configMEMORY_TAG_SIM_THRESHOLD
configMEMORY_TAG_LEVENSHTEIN_THRESHOLD
configMEMORY_LLM_PROBE_TTL_SEC60 Cache TTL for Ollama availability probe
configMEMORY_RECALL_EXCLUDE
configMEMORY_RECALL_EXCLUDED_TAGS
🔐 secretCOHERE_API_KEY
configMEMORY_LLM_API_BASEexport =https://openrouter.ai/api/v1
configMEMORY_EMBED_PROVIDERKeep =fastembed (default 384d) and only change the LLM provider, OR
configMEMORY_EMBED_API_BASE
🔐 secretMEMORY_EMBED_API_KEYEmbedding backends stay on fastembed by default. Switch via V9_EMBED_BACKEND=openai-3-large (set ) — costs ~$0.10/5k rows for re-embed, expected R@5 lift on conversational data.
configMEMORY_EMBED_MODELexport =text-embedding-3-small # 1536d
configMEMORY_ACTIVECONTEXT_DISABLEDisable: export =1
configV9_EMBED_BACKENDEmbedding backends stay on fastembed by default. Switch via =openai-3-large (set MEMORY_EMBED_API_KEY) — costs ~$0.10/5k rows for re-embed, expected R@5 lift on conversational data.
configV9_LOCOMO_TUNED_PATH
configV9_RERANKER_MODEL
configV9_RERANKER_FP16
configMEMORY_MODEfast default) and isolates install tests from real
configMEMORY_USE_LLM_IN_HOT_PATHfalse Master switch for sync LLM stages in save_knowledge / Recall.search. MEMORY_MODE=deep flips this to true.
configMEMORY_ALLOW_OLLAMA_IN_HOT_PATHWarm, in-memory SQLite, MacBook M-series, MEMORY_MODE=fast, =false:
configMEMORY_RERANK_ENABLEDfalse Honour caller's rerank=true. When false, CrossEncoder rerank is hard-disabled even if a tool call requests it.
configMEMORY_ENRICHMENT_ENABLEDfalse Run the async enrichment worker. Default-ON in balanced / deep.
configMEMORY_TEXT_EMBED_MODELsentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
configMEMORY_CODE_EMBED_MODELjinaai/jina-embeddings-v2-base-code # optional
configMEMORY_LOG_EMBED_MODELfalls back to TEXT
configMEMORY_CONFIG_EMBED_MODELfalls back to TEXT
configMEMORY_DEFAULT_EMBEDDING_SPACEtext Space for unclassified content.
configMEMORY_CONTRADICTION_TYPES
configMEMORY_CONTRADICTION_TOP_K
configMEMORY_CONTRADICTION_MIN_COSINE
configMEMORY_CONTRADICTION_LLM_THRESHOLD
configMEMORY_CONTRADICTION_FLAG_THRESHOLD
configMEMORY_CONTRADICTION_TIMEOUT_SEC
configMEMORY_CONTRADICTION_MODEL
configMEMORY_COREF_ENABLED
configMEMORY_COREF_HISTORY_LIMIT
configMEMORY_COREF_TIMEOUT_SEC
configMEMORY_COREF_MODEL
configMEMORY_ENRICH_TICK_SEC0.1 Worker tick interval (clamp 0.01..5)
configMEMORY_ENRICH_BATCH5 Rows claimed per tick (clamp 1..50)
configMEMORY_ENRICH_MAX_ATTEMPTS3 Retries before flipping a row to failed
configMEMORY_ENRICH_STALE_AFTER_SEC60 Seconds before a processing row is reclaimed (worker crash recovery)
configMEMORY_ENTITY_DEDUP_ENABLED
configMEMORY_ENTITY_DEDUP_THRESHOLD
configMEMORY_ENTITY_DEDUP_MAX_CANDIDATES
configMEMORY_ENTITY_DEDUP_TYPES
configMEMORY_ENTITY_DEDUP_LOG_ALL
configMEMORY_EPISODIC_MAX_ENTITIES_PER_EVENT
configMEMORY_GRAPH_EXPAND
configMEMORY_INBOX
configMEMORY_USE_BINARY_SEARCH
configMEMORY_OUTBOX_GRACE_SEC
configMEMORY_OUTBOX_MAX_ATTEMPTS
configMEMORY_WIKI_ENABLED
configMEMORY_WIKI_DIR
configMEMORY_WIKI_RECENT_DAYS
configMEMORY_WIKI_AUTO_REFRESH_EVERY_N
configMEMORY_QUALITY_THRESHOLD
configMEMORY_QUALITY_MIN_CHARS
configMEMORY_QUALITY_BYPASS_TYPES
configMEMORY_QUALITY_LOG_ALL
configMEMORY_QUALITY_TIMEOUT_SEC
configMEMORY_QUALITY_MODEL
configMEMORY_QUERY_REWRITE
configMEMORY_SMART_ROUTER
configMEMORY_SMART_ROUTER_FORCE
// full audit trail
The full breakdown of what we checked, the deductions that landed, the network hosts, the dependency advisories, and concrete fix guidance is available to verified publishers.
// improvement guidance — verified publishers only
We have 2 concrete improvements we can share with the publisher of this MCP. Each comes with specific guidance to raise the trust score.
// embed badge in your README
[![M8ven Score](https://m8ven.ai/badge/mcp/vbcherepanov-total-agent-memory-1xeo59)](https://m8ven.ai/mcp/vbcherepanov-total-agent-memory-1xeo59)
commit: 616d9a6f8b507c16b4cdfef4e823af59d949cc09
code hash: 477afea77b65c13242da653f01cc6432c844af79ababebf792f0dc8f62ab0acc
verified: 6/22/2026, 12:36:23 PM
view raw JSON →