LLM Kosh
Local-first AI memory cartridge with persistent MCP memory for Claude via SQLite FTS5.
Connector by rastogivaibhav · GitHub account verified by the MCP Registry
Chalk Labs
Need help deploying LLM Kosh?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Codex, VS Code, Cursor, Windsurf, Claude Desktop. Read the repository before you give any tool access to your data.
npx agentmarket install connector/rastogivaibhav-llm-kosh
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this server can do and what it asks for, from its tool list and install config. Checked Oct 25, 2026.
- Tools
- 23 read · 7 write · 0 delete · 6 not yet classified
- Secrets
- None asked for
- Install scripts
- Not scanned yet
- Runs as
- Python package
Tools by what they do 36
Read from the source code; the running server may list different tools.
Read 23 · Look things up without changing anything.
company_artifact_inspectcompany_artifact_segmentcompany_artifact_snapshotcompany_brain_healthcompany_episode_getcompany_episodes_searchcompany_memory_getcompany_memory_searchcompany_sessions_listget_cartridge_memory_mapget_daemon_statusget_project_context
Show 11 more tools
intake_convert_filekosh_verifylist_intakereasoning_explorereasoning_querysearch_memorytrusted_memory_conflictstrusted_memory_explaintrusted_memory_inboxtrusted_memory_recalltrusted_memory_review
Write 7 · Create, change, send or run things.
apply_intake_proposalcompany_artifact_registercompany_memory_proposecompany_memory_propose_from_evidencecompany_session_understandcreate_private_context_packsubmit_memory_receipt
Not yet classified 6 · We could not tell from the name and description.
company_brain_evaluatecompany_context_compilecompany_memory_reviewreasoning_critiquereasoning_ingesttrusted_memory_propose
Secrets and settings it asks for
Other settings (environment variables or headers).
CARTRIDGE_WORKSPACE
Files it touches
- Any path the agent passes to its tools
Tools (36)
Read from the source code; the running server may differ. Descriptions are the server’s own words; hidden characters are shown as [U+…].
apply_intake_proposalDirectly applies an intake proposal batch to the live memory.
company_artifact_inspectVerify and inspect a bounded region of an authorized registered artifact.
company_artifact_registerRegister an existing local artifact by fingerprint without copying its bytes.
company_artifact_segmentInspect an artifact and persist bounded derived segments with native citations.
company_artifact_snapshotExplicitly materialize an authorized artifact into immutable snapshot storage.
company_brain_evaluateRun reference storage, citation, and projection acceptance checks.
company_brain_healthValidate canonical company-brain storage and projection consistency.
company_context_compileCompile structured, cited, token-budgeted context for an agent task.
company_episode_getGet one authorized episode with ordered event and candidate provenance.
company_episodes_searchSearch authorized goal-oriented work episodes and observed outcomes.
company_memory_getGet one authorized memory with its evidence references and lifecycle.
company_memory_proposeCreate immutable evidence and a non-authoritative candidate memory.
company_memory_propose_from_evidenceCreate a candidate semantic memory citing an existing registered artifact.
company_memory_reviewApply a governed lifecycle transition to an authorized memory.
company_memory_searchPermission-first hybrid search over atomic, evidence-backed company memory.
Show 21 more tools
company_session_understandBuild normalized sessions, episodes, and cited candidates from JSONL evidence.
company_sessions_listList authorized normalized source sessions.
create_private_context_packCreates a highly focused context pack for a specific task. Includes private and sensitive information.
get_cartridge_memory_mapReturns a structural map of the knowledge base projects and active concepts.
get_daemon_statusVerifies the cryptographic ledger of the Cartridge.
get_project_contextFetches all context and active decisions for a specific project.
intake_convert_fileConverts a local file (e.g. PDF, DOCX, XLSX, PPTX, PNG, WAV) to structured markdown using the MarkItDown converter and ingests it into the cartridge.
kosh_verifyVerify a question against local temporal/causal evidence. This is read-only. The returned JSON report can include supporting facts, causal paths, contradictions, inferred-but-not-discovered relationships, missing evidence, stability information, and an explicit abstention state. It does not consult the internet, seed demo data, write or mutate memory, export private context, or make an imported source trustworthy.
list_intakeLists the current items pending in the intake queue.
reasoning_critiqueRun the Lyapunov critic on a specific list of fact IDs. Returns stability score, status, and per-dimension breakdown.
reasoning_exploreEnumerate all causal paths between two known facts. Returns the fiber bundle for that specific pair.
reasoning_ingestAdd a bounded atomic fact to the Temporal Causal Reasoning Graph. This is a write operation and requires --allow-write. 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_queryQuery 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_memorySearch the Cartridge Knowledge Base for memories, decisions, and files.
submit_memory_receiptSubmits a MEMORY_RECEIPT.md formatted string to the intake queue. Does not apply it directly; it must be reviewed and applied.
trusted_memory_conflictsList unresolved authorized Trusted Memory conflicts.
trusted_memory_explainExplain one authorized memory with evidence and admission history.
trusted_memory_inboxList authorized candidate and quarantined Trusted Memory items.
trusted_memory_proposePropose governed memory. Requires MCP write capability. If ``evidence_id`` is omitted, the proposal is persisted as an ``agent_observation``. Agents cannot self-label new MCP evidence as ``user_direct``. To preserve a stronger or different source type, register evidence first and pass its existing id.
trusted_memory_recallRecall governed memory with admission and lifecycle policy enforced.
trusted_memory_reviewExplicitly approve, quarantine, reject, or supersede governed memory. Requires the existing MCP mutate capability. Conflict resolution is never inferred from a generic approval.
Signals
- Stars
- 1
- Forks
- 1
- Open issues
- 0
- Weekly downloads
- 70
- Last commit
- Sep 14, 2026 (1 month ago)
- Last release
- Jun 23, 2026 (4 months ago)
- Last package publish
- Jun 23, 2026 (4 months ago)
From GitHub (fetched Oct 24, 2026) and the package registry.
Details
- Registry name
io.github.rastogivaibhav/llm-kosh- Version
- 2.1.3
- Auth
- No auth
- Packages
- pypi llm-kosh
- License
- MIT
- Repository
- github.com/rastogivaibhav/llm-kosh
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get LLM Kosh running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
LLM Kosh MCP server. Agent Mart. https://agentmart.dev/connectors/rastogivaibhav-llm-kosh (last updated 2026-10-24)