MCP.Market

LLM Kosh MCP server

by rastogivaibhav·io.github.rastogivaibhav/llm-kosh·v2.1.3·1 stars
No reviews yet

Local-first AI memory cartridge with persistent MCP memory for Claude via SQLite FTS5.

A92/100grade A
Write a reviewInstall
What users say
No reviews yet
Be the first
Safety scan
A92/100

full report

Adoption
Growing

1 stars27 downloads/wk

Reviews

Write one

Nobody has reviewed LLM Kosh yet.

If you have run it, two minutes of your experience saves the next person an afternoon.

LLM Kosh tools (13, 3 write)

write = sends, deletes, buys or posts

Read from the package source without running it. The installed server may list more.

  • apply_intake_proposal

    Directly applies an intake proposal batch to the live memory.

  • create_private_context_packwrite action

    Creates a highly focused context pack for a specific task. Includes private and sensitive information.

  • get_cartridge_memory_map

    Returns a structural map of the knowledge base projects and active concepts.

  • get_daemon_status

    Verifies the cryptographic ledger of the Cartridge.

  • get_project_context

    Fetches all context and active decisions for a specific project.

  • intake_convert_file

    Converts a local file (e.g. PDF, DOCX, XLSX, PPTX, PNG, WAV) to structured markdown using the MarkItDown converter and ingests it into the cartridge.

  • list_intake

    Lists the current items pending in the intake queue.

  • reasoning_critiquewrite action

    Run the Lyapunov critic on a specific list of fact IDs. Returns stability score, status, and per-dimension breakdown.

  • reasoning_explore

    Enumerate all causal paths between two known facts. Returns the fiber bundle for that specific pair.

  • reasoning_ingest

    Add a new fact to the Temporal Causal Reasoning Graph. documented_at and valid_from are ISO 8601 datetime strings. causal_edges: JSON array of {"target_id": str, "edge_type": str, "confidence": float}. Returns the new fact_id.

  • reasoning_query

    Query the Temporal Causal Reasoning Graph. By default returns a human-readable causal narrative (narrative=True). Set narrative=False to get raw JSON with full fiber bundle details. temporal_context: ISO 8601 datetime or Unix timestamp string (omit for now).

  • search_memory

    Search the Cartridge Knowledge Base for memories, decisions, and files.

  • submit_memory_receiptwrite action

    Submits a MEMORY_RECEIPT.md formatted string to the intake queue. Does not apply it directly; it must be reviewed and applied.

Public scan report

scanner v0.1.9 · 2026-09-26 · same rubric, same numbers if you re-run it

no findings
  • Code scan83 source files scanned25/25
  • –Live reliabilityno gateway calls yet and no remote to proben/a
  • –Tool poisoningtools not inspected (local package is not executed); not countedn/a
  • Auth qualitylocal package, no credentials required12/15
  • Maintenancelast push 13 days ago15/15
  • Maintainer identityregistry namespace matches repository owner; GitHub account older than a year8/10
Overall 92/100. Components that don't apply are left out of the denominator. Any critical finding is an F.RubricAppeal a findingJSON

How LLM Kosh compares

Against every graded Databases MCP server, from our own scans and uptime checks. Updated after each scan run.

Safety score
92/100
higher than 90% of 802 Databases servers
Tools
13
the typical Databases server has 8
Tools that change things
23%
typical: 10%. Fewer write tools means less an agent can break
GitHub stars
1
more than 45% of 602 Databases servers
Downloads a week
27
more than 13% of 328 Databases servers with a package

What its tools let an agent do: 9 read · 1 create · 2 change · 1 run or deploy

Install LLM Kosh in Claude Code, Cursor or VS Code

claude mcp add llm-kosh -- uvx llm-kosh
Add to Cursor

What the publisher says

From the LLM Kosh repository's README, as published. We do not edit it. Read it on GitHub

LLM-Kosh Local-first durable memory for AI agents. Give Claude, Cursor, and other MCP-compatible clients persistent, inspectable memory without handing your workspace to a hosted memory service. Kosh Verify makes that memory show its work: evidence, timing, causal paths, contradictions, inference boundaries, missing evidence, and abstention. Quickstart · Kosh Verify · Architecture · MCP · Security · Contributing

<!-- mcp-name: io.github.rastogivaibhav/llm-kosh -->

Why this exists

AI agents can reason across increasingly long workflows, but their memory is often either ephemeral or delegated to opaque hosted services.

LLM-Kosh treats memory as local infrastructure: inspectable, portable, auditable, permissioned, and usable across MCP-compatible clients.

It gives agents a durable memory layer built from ordinary local files plus structured indexes and governance controls:

  • Local-first — your cartridge stays on your machine by default.
  • Inspectable — memory remains readable, backupable, diffable, and reviewable.
  • Auditable — mutations are recorded in a tamper-evident ledger.
  • Permissioned — MCP starts read-only; write, mutation, and private export require explicit opt-in.
  • Portable — one cartridge can support multiple compatible AI clients and workflows.
  • Evidence-aware — Kosh Verify can distinguish support, contradiction, inference, evidence gaps, and absence.

Install it: python -m pip install --upgrade llm-kosh

Kosh Verify: memory that can show its work

Long-lived memory creates a different failure mode from a one-off bad answer: a weak or misunderstood memory can be recalled again in later sessions.

Kosh Verify is LLM-Kosh's evidence-aware verification surface. Given the evidence already present in a cartridge, it can produce a structured report containing temporal context, supporting facts, causal paths, contradictions, inferred-but-not-discovered relationships, missing evidence, stability information, and an explicit abstention state when there is not enough evidence.

It is not a universal truth oracle and it does not make an imported source trustworthy simply because it was stored. The aim is to preserve the difference between what the cartridge observed, what it inferred, what conflicts, and what it cannot support.

Try the deterministic synthetic incident demo:

llm-kosh --root ./kosh-demo kosh-verify \
  "Why did checkout fail and what evidence contradicts the explanation?" \
  --when "2026-05-01T13:30:00+00:00" \
  --depth 5 \
  --demo-seed \
  --json

Shortened. The full README is on GitHub.

Nothing above is checked by us. What we check is on the safety report.

LLM Kosh: common questions

Is LLM Kosh MCP server safe?
Yes, by our scan: it is graded A (92/100). Read the LLM Kosh safety report
How do I install LLM Kosh?
It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
Does LLM Kosh need an API key?
No secret keys are declared. It reads 1 setting from the environment.
Is LLM Kosh maintained?
The last commit was 51 days ago (2026-09-14). The latest release is v2.1.3.
What can I use instead of LLM Kosh?
Servers from other publishers that do the same job: Remem MCP server, open-memex MCP server and SAME - Stateless Agent Memory Engine MCP server. Compare all LLM Kosh alternatives.
How does LLM Kosh compare with other Databases MCP servers?
safety score 92/100 (higher than 90% of 802 Databases servers); tools 13 (the typical Databases server has 8); tools that change things 23% (typical: 10%. Fewer write tools means less an agent can break). Full comparison.

Alternatives to LLM Kosh

Same job from other publishers: the closest match first, then the best rated.

All LLM Kosh alternatives →