Factsets MCP Server

Read from the package source without running it. The installed server may list more.

• add_resources

  Register one or more resources (files, URLs, APIs, commands) with metadata for tracking. The system stores retrieval methods - actual fetching is performed by you. Register immediately after fetching - don't wait for user prompting.

• build_skill_context

  Build formatted context for a skill including its full content, tags, references, and optionally hydrated referenced skills. Returns markdown suitable for agent consumption.

• check_stale

  Check for stale resources, skills with changed dependencies, and unverified facts. Returns retrieval methods for stale resources so you can refresh them.

• create_skillwrite action

  Create a new skill document. Skills are markdown files that capture procedural knowledge and can reference other skills, resources, and facts.

• create_tagswrite action

  Create one or more tags for organizing facts, resources, and skills

• delete_configwrite action

  Delete a configuration value by key

• delete_factswrite action

  Delete facts by ID, tags, age, or verification status

• delete_resourceswrite action

  Delete resources by ID or URI. Also removes associated tags and skill references.

• delete_skillswrite action

  Delete skills by name. Optionally delete the skill files from disk.

• get_config

  Get a configuration value by key

• get_config_schema

  Get the configuration schema showing all available options with descriptions, types, and defaults. Use this to discover configurable settings like freshness thresholds.

• get_dependency_graph

  Get a dependency graph for a skill showing all connected skills, resources, and facts. Returns a tree structure with nodes and edges for visualization. Use this to fully saturate skill context before performing complex tasks.

• get_execution_log

  Get a specific execution log by ID to see full details including output.

• get_knowledge_context

  Build a knowledge context from tags. Returns formatted markdown with relevant facts, resources, skills, and staleness warnings. Use this to gather context before working on a task.

• get_maintenance_report

  Generate a maintenance report showing stale resources, skills with changed dependencies, and unverified facts. Use this to identify knowledge base items that need attention.

• get_preference_prompt

  Generate a natural language prompt describing all user preferences. Use this to understand how to format responses, code, and documentation according to user preferences.

• get_refresh_guide

  Get step-by-step instructions for refreshing a specific resource. Returns guidance on how to fetch new content based on the resource's retrieval method.

• get_resource

  Get a specific resource by ID or URI. Returns the stored snapshot and retrieval method for you to refresh if needed.

• get_resources

  Retrieve multiple resources by IDs or URIs in a single call. Returns found resources and lists any not found.

• get_skill

  Retrieve a skill by name with its full content, tags, and references. Set hydrateRefs to include content of referenced skills.

• get_skills

  Retrieve multiple skills by name in a single call. Returns found skills and lists any not found.

• get_user_preferences

  Get all user preferences including communication style, code output, documentation, and interaction settings

• infer_preferencewrite action

  Update a user preference based on explicit statement or inferred behavior. Use when user indicates a preference (e.g., 'no emojis', 'be more concise'). Explicit preferences are always applied; inferred ones require high confidence.

• link_skill

  Add references from a skill to other skills, resources, or facts

• list_config

  List all configuration values with their current settings

• list_tags

  List all tags in the knowledge base, optionally filtered by name pattern

• mark_resources_refreshed

  Mark resources as refreshed after you have verified their content is still valid

• prune_orphan_tags

  Clean up orphan tags that have zero usage (not linked to any facts, resources, skills, or execution logs). Use dryRun=true to preview which tags would be deleted without actually deleting them.

• reset_preferences

  Reset user preferences to defaults. Optionally specify keys to reset specific preferences only.

• restore_facts

  Restore soft-deleted facts by their IDs. Returns the count of facts that were restored.

• restore_resources

  Restore soft-deleted resources by their IDs. Returns the count of resources that were restored.

• restore_skills

  Restore soft-deleted skills by their names. Returns the count of skills that were restored.

• search_execution_logs

  Search execution history by tags, free text query, or success status. Find what commands were run, what worked, and learn from past experiments. Query searches across command, context, and output fields.

• search_facts

  Search for facts by tags, content query, verification status, or source type. Use this on every prompt to validate context before responding - not just at session start.

• search_resources

  Search for resources by tags, type, or URI pattern. Check for existing resources before registering new ones to avoid duplicates.

• search_skills

  Search for skills by tags or title query. Check for existing skills before creating new ones to avoid duplicates.

• set_config

  Set a configuration value. Use get_config_schema to see all available options with descriptions and valid values. Freshness settings control staleness thresholds in hours.

• submit_execution_logswrite action

  Record execution logs for commands, tests, builds, or any actions. Captures command, output, success status, and tags for future reference. Use this to build institutional memory of what commands work for this project. Log immediately after running commands - don't batch or delay.

• submit_factswrite action

  Submit one or more facts to the knowledge base. Facts with matching content will be updated. Call this immediately when learning something new - do not wait for user prompting.

• sync_skill

  Sync a skill's content hash after the file has been modified externally. Reads the skill file and updates the stored hash if changed.

• update_factwrite action

  Update an existing fact's content, metadata, source, verification status, or tags. Find the fact by ID or exact content match. Supports replacing all tags, appending new tags, or removing specific tags.

• update_resourcewrite action

  Update resource metadata (description, tags, retrieval method) without modifying snapshot content or lastVerifiedAt timestamp. Use this to fix placeholder descriptions or reorganize tags.

• update_resource_snapshotwrite action

  Update the snapshot content for a resource after you have fetched fresh data using the retrieval method

• update_resource_snapshotswrite action

  Bulk update snapshot content for multiple resources. Each entry requires resourceId and snapshot content.

• update_skillwrite action

  Update an existing skill's metadata, tags, or references. Does not modify skill file content - use sync_skill after editing the file directly.

• update_tagswrite action

  Update descriptions for existing tags. Use this to replace auto-generated placeholder descriptions with meaningful content.

• verify_facts

  Mark one or more facts as verified/confirmed

• verify_facts_by_tags

  Bulk verify all facts that have specified tags. By default, verifies facts with ANY of the tags. Set requireAll=true to only verify facts that have ALL specified tags. Returns the count and IDs of verified facts.

• Code scan202 source files scanned20/25
• –Live reliabilityno gateway calls yet and no remote to proben/a
• –Tool poisoningtools not inspected (local package is not executed); not countedn/a
• Auth qualitylocal package, no credentials required12/15
• Maintenancelast push 201 days ago4/15
• Maintainer identityregistry namespace matches repository owner; GitHub account older than a year8/10

Safety score

68/100

higher than 30% of 48,255 servers

Tools

48

the typical server has 7

Tools that change things

31%

typical: 0%. Fewer write tools means less an agent can break

GitHub stars

1

more than 65% of 28,688 servers

Downloads a week

155

more than 71% of 12,191 servers with a package

Factsets MCP Server: common questions

Is Factsets MCP Server safe?

With care: it is graded C, so read the findings first (68/100). Read the Factsets MCP Server safety report

How do I install Factsets MCP Server?

It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.

Does Factsets MCP Server need an API key?

Not as far as the registry entry and our scan can tell: no credentials are declared or required.

Is Factsets MCP Server maintained?

The last commit was 235 days ago (2026-03-14). The latest release is v0.2.0.

How does Factsets MCP Server compare with other MCP servers?

safety score 68/100 (higher than 30% of 48,255 servers); tools 48 (the typical server has 7); tools that change things 31% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.