# SilvaLabs Muse connector

From musedirectory.ai, the independent directory of Meta Muse connectors. Not affiliated with Meta.

## SilvaLabs

Lets Muse read only the personal notes and documents you choose to open.

- Record: https://musedirectory.ai/connector/silvalabs
- Category: Productivity
- Developer: Silva Labs (https://silvalabs.app/en/for-agents/)
- Muse status: Extra setup. Not in Muse's Connectors list yet. Muse can still use it: its page gives you a request to paste into Muse.
- Health: Working, 92ms, checked 2026-11-20T15:46:30Z
- Endpoint: https://mcp.silvalabs.app/mcp
- Auth: Needs an access key; Pricing: unknown
- Screening: Screened, needs a closer look (2026-11-20T16:26:20Z)
- Source: Found in the official MCP Registry (io.github.opensylva/silvalabs) https://registry.modelcontextprotocol.io/v0/servers?search=io.github.opensylva%2Fsilvalabs

Connects Muse to a personal knowledge base where you control which materials agents can read. Muse can search and reference the documents you open while the rest stays private.

Example request: "Look through the notes I opened in SilvaLabs and summarize what I decided about the kitchen renovation."

How to connect: Not in Muse's Connectors list yet, but Muse can still use it. Paste this into Muse: "Use SilvaLabs to help me. It is a free service with an MCP server at https://mcp.silvalabs.app/mcp. It needs an API key from Silva Labs; ask me to enter it through your secure credential prompt. Ask me before you share anything with it." Muse asks before it shares anything with the app's site. Meta does not review apps used this way, so only use ones you trust. We tested this in the Muse app on September 24, 2026: Muse used an app's link directly this way and returned a live answer.

Tools:
- list_collections: When to use: folder tree / find which folder exists. Not for docs inside a folder (get_index), not for keyword search (search_docs), not for a pasted workset (get_workset). List the Silvalabs folder tree for the agent: every node on the path to a leaf that has at least one visibl
- get_index: When to use: list docs in one known folder (or unfiled). Not for folder tree (list_collections), not for keyword search across folders (search_docs), not for a workset id (get_workset). Table of contents for one folder: only docs with visible_to_agent=true (hidden cards do not ap
- get_workset: When to use: only with a workset UUID / silvalabs:workset/… paste. Not a folder (get_index), not folder tree (list_collections), not keyword search (search_docs). Flat list of docs in one Silvalabs workset (a temporary task selection, not a folder): only docs with visible_to_agen
- get_doc: Read one document: metadata + compact brief/TOC by default to save context. Do NOT use view='full' unless you specifically need the complete raw text or code. Prefer reading specific sections via get_doc_section using the chunk_index or heading from toc. Default view (omit or hea
- create_doc: Create a new Silvalabs library card from the agent (needs API key scope ingest:write). Accepts markdown text OR a http(s) URL — not both. Files/PDF/DOCX uploads are NOT supported; if you have a file, save a link (url) instead. Video/audio URLs (YouTube, Vimeo, media files) are sa
- update_agent_md: Update working agent notes (artifact agent_md) only on cards created by the agent (source_type=api). Human url/file cards → forbidden. Does NOT replace the source text (clean_md). Needs ingest:write. Pass id as UUID or silvalabs:doc/<uuid>/…. markdown empty string clears agent_md
- update_doc_body: Replace the document body (clean_md) only on cards created by the agent (source_type=api). Human url/file cards → forbidden. Needs ingest:write. PUT /api/v1/source-items/{id}/agent/clean-md base_fingerprint is REQUIRED — take clean_md_fingerprint from get_doc (OCC; mismatch → CON
- delete_doc: Write: soft-delete an agent-owned doc/card (source_type=api) only. Human url/file cards → forbidden via MCP. After success the doc is no longer visible to the agent (soft-delete, not a hard wipe). DELETE /api/v1/source-items/{id}/agent confirm_title must exactly match the effecti
- get_doc_section: Read one section of a document from the index (FULL_TEXT_CHUNKS). Use chunk_index or heading from get_doc toc — prefer this over view=full. heading match is fuzzy (case-insensitive; leading # ignored; path leaf / segment). If the index is empty: not found — wait for indexing or u
- search_docs: When to use: keyword/hybrid search when the folder is unknown or the question spans folders. Not for browsing one known folder (get_index), not for folder tree (list_collections), not for a workset list (get_workset). Prefer list_collections → get_index when the folder is known. 
- get_procedure: Read-only: procedure overview — status, current step summary, next_action (needs scenario:execute). Not a single-step body (use view_step). Never invent attempt_id; follow next_action. Without grant you may read but cannot claim; with grant continue via take_step / resume_step.
- view_step: Read-only: current or given procedure step plus previous_attempts; follow next_action. Omit step_id for the current step. Pass attempt_id only for your own attempt. Does not claim or mutate; use take_step / complete_step / block_step to change state.
- take_step: Write: claim a WAITING_AGENT step (needs grant). Returns attempt_id; follow next_action. For an already RUNNING step use resume_step. Do not invent attempt ids. Optional expected_*_revision for optimistic concurrency (stale → conflict).
- resume_step: Write: resume or take over a RUNNING step (needs grant). Returns attempt_id; follow next_action. Set confirm_takeover=true only for another agent's RUNNING step. For WAITING_AGENT use take_step. Optional expected_*_revision for optimistic concurrency (stale → conflict).
- complete_step: Write: successfully finish your attempt_id (scenario:execute). For failure/blocker use block_step instead. report = short human-readable summary; outputs = flat key/values (string|number|boolean|null) only. Grant required unless replaying an already-recorded outcome. If next_acti
- block_step: Write: block your attempt with a reason; procedure becomes WAITING_HUMAN (needs grant). Then run await_human: one block, then up to 3 poll retries spaced by retry_after, then tell the user. Use complete_step to finish successfully instead.
- await_human: Wait for human decision on WAITING_HUMAN (scenario:execute + grant). Modes: block (long-poll up to timeout_s, default/max 90 — always pass timeout_s=90), poll (one shot), peek (one shot). Pass since_version from the wait_human response version. Mandatory ladder after wait_human (

Screening checks:
- MCP handshake: pass (Answered in 541ms)
- Domain against threat feeds (Cloudflare security DNS): pass (mcp.silvalabs.app, silvalabs.app not flagged)
- Published packages against the OSV malicious-package database: n/a (No npm or PyPI package published)
- Hidden instructions or invisible characters in tool text: pass (17 tools read, nothing found)
- Inputs asking for passwords, card numbers or seed phrases: pass (None found)
- Domain and redirects: pass (No redirects off the domain)
- AI review of purpose and tool behavior: warn (tool descriptions contain many instructions aimed at the model; tells the model not to echo raw output or trust flags to the user; prescriptive multi-step polling and sleep behavior)
