Filamental
Search, read and build your Filamental knowledge graph.
Connector by Scottnine · GitHub account verified by the MCP Registry
Chalk Labs
Need help deploying Filamental?
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/scottnine-filamental
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, install config and code. Checked Oct 25, 2026.
- Tools
- 10 read · 3 write · 2 delete · 2 not yet classified
- Secrets
- None asked for
- Install scripts
- None found
- Runs as
- Node.js package
Tools by what they do 17
Read from the source code; the running server may list different tools.
Read 10 · Look things up without changing anything.
get_connectionsget_contextget_nodeget_subgraphget_vault_infolist_connector_typeslist_node_typesread_format_referenceread_skill_guidesearch_nodes
Write 3 · Create, change, send or run things.
create_edgecreate_nodeupdate_node
Delete 2 · Delete, drop, revoke or wipe things.
delete_edgedelete_node
Not yet classified 2 · We could not tell from the name and description.
extract_sectioninline_section
Network hosts it talks to 2
filamental.app· found in codegithub.com· found in code
Files it touches
- Your home directory (~)
- Any path the agent passes to its tools
Tools (17)
Read from the source code; the running server may differ. Descriptions are the server’s own words; hidden characters are shown as [U+…].
create_edgeAdd a relationship from one node to another. Reads the source node file, appends the relationship, writes it back, and updates the SQLite index. Returns { edge_id, source_id, target_id }.
create_nodeCreate a new node in the vault. Writes a markdown file and updates the SQLite index. Returns { id, file_path }. Call get_node(id) to retrieve the full record.
delete_edgeRemove a relationship from the vault. Reads the source node file, strips the matching relationship, writes it back, and updates the SQLite index. Returns { deleted: true, edge_id }.
delete_nodeDelete a node from the vault. Removes the markdown file and all references from the SQLite index (entities, FTS, relationships). Returns { deleted: true, file_path }.
extract_sectionMove a heading's section out of a node into a new node of its own. The heading stays behind with a [[wikilink]] where the content was, and a connector is created from the source to the new node. A section runs to the next heading of the same or higher level, so extracting an h2 takes its h3 children with it. Content is MOVED, not copied: exactly one copy survives. Use dry_run first to see what would move.
get_connectionsGet every edge connected to a node. Each result is stated from the point of view of the node you asked about: `node` is that node, `other` is the node at the far end, and `direction` says where the arrowhead is drawn as seen from `node` — "outgoing" points away at `other`, "incoming" points back at `node`, "bidirectional" points both ways, "none" is a plain line with no arrow. This is what the user actually sees on the graph. It is deliberately not the raw stored value, which is relative to whichever end the connector happened to be drawn from and is invisible to the user. `stored_on` names th…
get_contextAssemble one readable markdown document for a node and its neighbourhood: the node's notes plus the notes of every node within `depth` hops, inlined with headings that track hop distance and the connector each was reached by. Use this instead of get_subgraph + repeated get_node when you want to READ the surrounding content rather than inspect graph shape. Set dry_run to size the result before pulling it.
get_nodeRetrieve full node data for a given entity UUID, including the markdown note body in the `notes` field.
get_subgraphBFS traversal from a root node collecting all reachable nodes and edges up to the given hop depth. Results are deduplicated.
get_vault_infoReturn summary counts (nodes, edges) and the top-level entity and connector type names configured for this vault.
inline_sectionFold a node's content back into another node: the inverse of extract_section. The content lands where the [[wikilink]] to it already sits, or is appended if there is no such link, and the connector between the two is removed. The target node is DELETED by default, because keeping it would leave the same text in two files. Deletion is refused if the target is connected to anything other than the source. Pass delete_target: false to keep the node as an empty stub instead. Use dry_run first.
list_connector_typesReturn the full connector type configuration for this vault (from .filamental/connector_types.json).
list_node_typesReturn the full entity type configuration for this vault (from .filamental/entity_types.json).
read_format_referenceRead the exact on-disk file format: the YAML frontmatter fields of a node markdown file, how relationships are stored, and what the .filamental folder holds. Needed only when reading or writing vault files directly rather than through these tools.
read_skill_guideRead the full Filamental skill guide: how to design a space, choose entity and connector vocabularies, set colours, arrow directions and physics weights, and the conventions that make a structure readable. The server instructions already carry the essentials, so reach for this when you need depth: building a substantial structure from scratch, or a question the essentials do not settle.
Show 2 more tools
search_nodesFull-text search across entity names, note bodies and property values. Returns matching nodes with a contextual snippet. An optional `filter` narrows results by field value, and may be used on its own to list nodes structurally without any search terms.
update_nodeUpdate an existing node. Only supplied fields are changed; omitted fields retain their current values. Providing relationships or properties replaces the entire array/map. Returns { id, file_path }.
Signals
- Stars
- 2
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Sep 4, 2026 (1 month ago)
- Last release
- Oct 26, 2026 (5 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- Registry name
io.github.Scottnine/filamental- Version
- 0.2.16
- Auth
- No auth
- Packages
- npm filamental-mcp
- License
- License unknown
- Repository
- github.com/Scottnine/filamental
- First seen
- Oct 24, 2026
- Last checked
- Oct 26, 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 Filamental 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
Filamental MCP server. Agent Mart. https://agentmart.dev/connectors/scottnine-filamental (last updated 2026-10-26)