cic-mcp-session
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@cic-mcp-sessionsearch session for 'deploy' status"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
cic-mcp-session
Session-scope esemény-/idővonal-/chunk-tár és provenance réteg a CIC agent-kontextus (cic-mcp-* család) számára.
A cic-mcp-* család trust-domain rétegezésében ez a komponens egyetlen beszélgetés/session
kontextusát tárolja és szolgálja ki — nem canonical tudás, nem cross-session memória.
Mi ez és mi nem
Igen:
SessionIngressEnvelopeingestraw event store
turn/timeline projection
chunk store
source/provenance refs
metadata index, full-text search, vector search
session-scope context pack
stabil SQL/API/MCP read tools
Nem:
canonical tudás
shared memory
cross-session graph
végleges döntésbányászat
human review nélküli promotion
A komponensek közti pontos határt lásd: CLAUDE.md.
Related MCP server: LatentContext MCP Server
Státusz
experimental — a repo a cic-mcp-factory job-lifecycle-én keresztül épül fel, kapacitás-jobonként.
A teljes session-specifikus adatfolyam (ingress envelope, Postgres write-path, projection,
chunk/embedding indexelés, FTS/vektor/hibrid retrieval, MCP read tool-ok) MÁR megépült és valódi
Postgres ellen bizonyítva van (~17 capability-job, lásd output/session-*-report.md):
SessionIngressEnvelopewrite-path:session_store/envelope_writer.py:165(insert_envelope) —output/session-raw-event-store-report.mdvalódi hook-producer:
hooks/log-event.py:303-304(insert_envelope()-et hív) —output/session-hook-collector-report.mdraw event store:
session_raw.envelopes(output/session-postgres-schema.sql) — ugyanaz a riportworker loop (turn projection + chunk indexelés, ütemezve):
session_store/turn_projector.py:300(run_projection_batch),session_store/chunk_indexer.py:378(run_indexing_batch),session_store/worker_loop.py:65,93(run_one_iteration/run_loop) —output/session-turn-projector-report.md,output/session-chunk-indexer-report.md,output/session-worker-scheduler-report.mdFTS / vektor / hibrid keresés:
session_api.search_context(),search_context_vector(),search_context_hybrid()(output/session-retrieval-quality-migration.sql,output/session-vector-search-api-migration.sql,output/session-hybrid-search-api-migration.sql) —output/session-vector-search-api-report.md,output/session-hybrid-search-api-report.mdsession_apiréteg (timeline, context pack, status, source refs):output/session-postgres-schema.sql+output/session-source-refs-api-migration.sql—output/session-source-refs-api-report.md7 tool-os MCP szerver:
mcp-server/session_server.py(search_session_context+search_session_context_fts/search_session_context_vector/get_session_timeline/get_session_context_pack/get_session_status/get_session_source_refs) —output/session-mcp-tools-report.md,output/session-mcp-tools-remaining-report.mdhost-natív MCP indítás (
.venv-host, nem a Docker-builderp_venv):.mcp.json.tpl—output/session-mcp-venv-fix-report.md
A dokumentált korlátok: a fenti komponensek production reachability-je jelenleg scaffold
szintű — a write-path-ot és a worker-loop-ot semmilyen permanens, deployolt infrastruktúra
(cron/systemd timer) nem hívja élesben (lásd output/session-worker-scheduler-report.md
"Risks"), és a cic-session MCP szerver nincs bekötve semelyik éles orchestrátor/Claude Code
session .mcp.json-jába (lásd output/session-mcp-config-wiring-report.md). A
status: experimental ezt a reachability-rést jelzi, NEM a session-specifikus implementáció
hiányát.
Kapcsolódó dokumentáció
cic-mcp-factoryfactory-docs — a komponens tervezési alapja (architecture.md,acceptance-contract.md,execution-phases.md)cic-mcp-knowledge— a canonical réteg, amire ez a komponens sosem promote-ol automatikusan
This server cannot be deployed
Maintenance
Related MCP Connectors
Hosted MCP memory and agent control plane for durable conversations, jobs, and operations.
Intelligent context infrastructure for AI teams: knowledge graph, sessions, tasks, documents.
Shared, versioned context that humans and AI agents can publish, review, annotate, and continue.
Persistent memory for AI agents — log and recall conversation context over MCP.
Related MCP Servers
- AlicenseAqualityDmaintenanceProvides persistent session memory for AI assistants, enabling them to store, search, and retrieve conversation summaries across sessions via the Model Context Protocol.10MIT
- AlicenseAqualityBmaintenanceA session-scoped memory layer for LLMs that enables AI assistants to explicitly store and retrieve notes, decisions, and context within a single conversation, ensuring focus without cross-session data contamination.65 npm5MIT
- FlicenseNot gradedqualityCmaintenanceProvides session persistence, crash recovery, decision tracking, and context compression for AI assistants, enabling seamless multi-session continuity.-
- AlicenseNot gradedqualityDmaintenanceEnables MCP-compatible AI clients to store and recall voice conversation context by caching user and AI utterances, supporting formatted context summaries for multi-turn voice interactions.MIT