Connection check
local install
anmol-srv-sigil
Persistent memory for AI coding agents. Local-first. MCP-native. Auto-captured from Claude Code, shared with Cursor, Codex, and any MCP client. Your agents don't remember what you decided yesterday. Sigil does. Every prompt, every session, your context is already there.
Tools
11
GitHub stars
10
Installs / wk
15
Licence
MIT
Transport
stdio
Last checked
never
Tools & capabilities
11 toolsRead from the published package source — this server runs locally, so there is no endpoint to query. Names are taken from the code, not observed at runtime, and descriptions are often absent.
get_entity_context
from source
Get full context for an entity: relations, connected facts, graph position. Use for: Get full context for an entity: relations, connected facts, graph position. Use for:
get_fact_context
from source
Get full context for a specific fact: complete content, entities mentioned, relations created, sources. Use for: drilling down on a fact from search results, checking provenance, u… Get full context for a specific fact: complete content, entities mentioned, relations created, sources. Use for: drilling down on a fact from search results, checking provenance, understanding connections. This is the detail view \u2014 search returns truncated facts, this returns everything.
get_pod
from source
Get a memory pod with its member facts, documents, and (for person pods) the canonical entity link. Use for: Get a memory pod with its member facts, documents, and (for person pods) the canonical entity link. Use for:
ingest
from source
Ingest a document into the Sigil knowledge base. Accepts raw content, a file path, or a URL. Parses the content, chunks it, embeds it, extracts facts, links entities, and stores ev… Ingest a document into the Sigil knowledge base. Accepts raw content, a file path, or a URL. Parses the content, chunks it, embeds it, extracts facts, links entities, and stores everything for search. Use when: adding documents to the knowledge base, ingesting files, URLs, or raw text.
list_pods
from source
List memory pods (typed containers segregating facts by source or subject). Use for: List memory pods (typed containers segregating facts by source or subject). Use for:
prime
from source
Prime your memory for this session \u2014 CALL THIS FIRST, before answering anything.\nThis client has no automatic memory: nothing is recalled or saved for you.\nReturns: Sigil he… Prime your memory for this session \u2014 CALL THIS FIRST, before answering anything.\nThis client has no automatic memory: nothing is recalled or saved for you.\nReturns: Sigil health (DB/providers) + fresh facts about the user and this project.\nIf you skip it, you start with zero knowledge of who the user is or what they
remember
from source
Save one or more standalone facts to the Sigil knowledge base. Each fact is classified, embedded, deduped against existing memory, and stored. Use when: the user states a durable p… Save one or more standalone facts to the Sigil knowledge base. Each fact is classified, embedded, deduped against existing memory, and stored. Use when: the user states a durable preference, decision, constraint, or factual claim worth recalling later. Pass each distinct fact as its own array element \u2014 don
search
from source
Search Sigil knowledge base for facts across all ingested documents. Automatically detects entity names and returns entity-centric results. Use for: Search Sigil knowledge base for facts across all ingested documents. Automatically detects entity names and returns entity-centric results. Use for:
search_entity
from source
Find entities in the knowledge graph by name or type. Use for: Find entities in the knowledge graph by name or type. Use for:
status
from source
Show Sigil knowledge base statistics \u2014 documents, chunks, facts, entities, relations. Use when: checking system health, verifying ingestion, reviewing knowledge graph size. Show Sigil knowledge base statistics \u2014 documents, chunks, facts, entities, relations. Use when: checking system health, verifying ingestion, reviewing knowledge graph size.
traverse_graph
from source