fini

Descriptor

Candidate descriptor · transport stdio

Tools

• assignKnowledgeToAgents — Assign or unassign folders to agents in bulk.
• bulkDeleteConversations — Delete up to 50 conversations in one request through the public Conversations API.
• bulkGenerateKnowledge — Queue generate-and-save jobs for multiple source IDs in one request.
• checkKnowledgeJobs — Check status for one or more background knowledge-generation jobs.
• crawlLinks — Discover additional web URLs from one or more seed links before ingesting them as sources.
• createArticle — Create a live article or draft article directly in the workspace.
• createArticleDraft — Create a draft from an existing live article.
• createKnowledgeFolder — Create a folder in the knowledge tree.
• createRule — Create a published rule.
• createRuleDraft — Generate a draft rule from instructions.
• createTag — Create a tag inside one tag group.
• createTagGroup — Create a custom tag group.
• deleteArticle — Delete an article.
• deleteConversation — Delete one conversation by ID through the public Conversations API.
• deleteKnowledgeFolder — Delete a folder from the knowledge tree.
• deleteRule — Delete one rule.
• deleteSources — Delete source records and optionally delete the linked Articles in the same request.
• deleteTag — Delete one tag by tag ID.
• deleteTagGroup — Delete a custom tag group.
• duplicateRule — Create a copy of an existing rule.
• generateAnswer — Send a message event into Fini and return the public events created for that submission.
• generateKnowledge — Queue one knowledge-generation job from candidate text, with optional source or inbox linkage.
• getArticlesByIds — Fetch one or more articles by ID.
• getConversation — Fetch one public conversation by ID.
• getKnowledgeFolders — Return the current knowledge-tree snapshot, optionally scoped to one agent.
• getPromptHistory — List saved prompt versions for one agent, newest first.
• getPrompts — Fetch the current merged prompts for one agent.
• getRule — Fetch one rule by ID with its full tree.
• getSource — Fetch one source record by ID.
• getTag — Fetch one tag by ID.
• getTagGroup — Fetch one tag group by ID.
• ingestSources — Queue first-time ingestion or refresh jobs for web links or existing source IDs.
• initializeKnowledgeFolders — Generate a downloadable tree-import template file from an initialization prompt.
• listAgents — List the agents in the workspace tied to your API key so you can map names to bot IDs.
• listArticles — List live articles, draft articles, or both.
• listConversations — Read all Fini conversations for your workspace, sorted newest first, with cursor pagination.
• listProviderResources — Discover importable resources from a connected provider. Step 1 of the provider import flow.
• listRules — List rules in the workspace by type and draft state.
• listSources — Paginate through every source record in the workspace, with filters on source type, title, URL, linked article, and linked operation.
• listTagGroups — List the tag groups visible to your workspace API key, including Fini defaults and workspace-owned custom groups.
• listTagsAcrossGroups — List tags across one or more tag groups as a flat array.
• listTagsInGroup — List the tags inside one tag group.
• moveArticle — Move an article into a different folder.
• moveKnowledgeFolder — Move a folder under a different parent in the knowledge tree.
• persistKnowledgeFolders — Upload a tree file and persist it into the workspace knowledge graph.
• publishArticleDraft — Publish a draft article.
• publishRuleDraft — Publish a draft rule and assign it to one or more agents.
• refineRuleDraft — Regenerate a draft rule using refinement instructions.
• refreshSources — Re-fetch existing source IDs so Fini can detect upstream changes and update linked knowledge.
• registerProviderResources — Create or update source records from provider resources. Step 2 of the connected-source flow.
• updateArticle — Update an existing article.
• updateKnowledgeFolder — Update a folder's title, description, or active state.
• updatePrompts — Save a new prompt version for one agent.
• updateRule — Update an existing rule.
• updateTag — Update one tag by tag ID.
• updateTagGroup — Update a custom tag group.

About MCP

The Model Context Protocol (MCP) is an open protocol Anthropic introduced for connecting LLM-based agents to external tools and data sources. Providers publish MCP servers that expose their API surface as structured, discoverable tools — an MCP-compatible client (Claude Desktop, Cursor, Cline, Continue, etc.) can connect to the server and call its tools without any per-provider integration code.

Browse every MCP server on the APIs.io network or compare with the broader Agent Skill surfaces of the same providers.