{"tool1":{"_id":"69b2da6867df398baec12e46","name":"Salesforce","slug":"salesforce","url":"https://salesforce.com","description":"","logo":"","category":"CRM","tags":[],"pricingModel":"unknown","scores":{"tokenEfficiency":{"score":6,"confidence":"scanner","evidence":"Salesforce APIs support field selection and pagination, but the platform's complex object model and mandatory fields often result in larger-than-necessary responses; no evidence of compact response formats or batch optimization.","na":false},"access":{"score":7,"confidence":"scanner","evidence":"Salesforce provides REST API, CLI (@salesforce/cli), and SDKs in Node and Python, plus structured data on homepage, but lacks GraphQL support and no MCP server integration, limiting modern agent frameworks.","na":false},"auth":{"score":7,"confidence":"scanner","evidence":"Salesforce supports OAuth 2.0 and API key authentication with scoped permissions, enabling autonomous agent authentication, though OAuth flow adds complexity compared to simpler API key-only systems.","na":false},"speed":{"score":6,"confidence":"scanner","evidence":"Salesforce has reasonable rate limits and API versioning, but no response time data collected, no evidence of ETag support for conditional requests, and concurrent request handling is standard but not explicitly optimized.","na":false},"discoverability":{"score":5,"confidence":"scanner","evidence":"Salesforce has developer documentation and structured data on homepage, but no OpenAPI spec found, no .llms.txt or agents.json files, making API discovery and integration harder than API-first platforms.","na":false},"reliability":{"score":6,"confidence":"scanner","evidence":"Salesforce supports API versioning and has mature infrastructure, but no evidence of idempotency key support or consistent error schemas; robots.txt blocks agents suggesting potential reliability concerns for automated access.","na":false},"safety":{"score":5,"confidence":"scanner","evidence":"Salesforce offers scoped OAuth tokens and sandboxes for testing, but no dry-run capability, no explicit undo mechanisms, and robots.txt blocking agents suggests safety-first design may limit autonomous operation.","na":false},"reactivity":{"score":5,"confidence":"scanner","evidence":"Salesforce supports webhooks and streaming APIs for some use cases, but no evidence of SSE, push notifications, or efficient polling mechanisms; webhook coverage may be incomplete across all object types.","na":false}},"agentGrade":"B","agentScore":6.1,"accessMethods":{"restApi":true,"graphql":false,"cli":true,"sdk":["Node (@salesforce/plugin-settings)","Python (salesforce)"],"mcpServer":"none","openApiSpec":"","llmsTxt":false,"agentsJson":false},"authInfo":{"methods":["unknown"],"scopedPermissions":false,"humanRequired":true},"reviewCount":0,"avgReviewScore":0,"viewCount":22,"badgeEmbedCount":2,"agentSkillSlugs":[],"alternatives":[],"claimed":false,"status":"graded","createdAt":"2026-03-12T15:23:20.671Z","updatedAt":"2026-04-09T03:06:55.192Z","__v":0,"scannerData":{"lastScannedAt":"2026-03-12T19:28:04.232Z","scanVersion":1,"rawSignals":{"homepage":{"status":200,"contentLength":598305,"hasStructuredData":true,"hasDeveloperDocs":true,"hasAgentMentions":true,"responseTimeMs":null},"openapi":{"found":false},"wellKnown":{"llmsTxt":{"found":false},"agentsJson":{"found":false},"robotsTxt":{"found":true,"blocksAgents":true,"hasSitemap":true}},"packages":{"npm":[{"name":"@salesforce/cli","description":"The Salesforce CLI","version":"2.126.4"},{"name":"@salesforce/plugin-settings","description":"configure the Salesforce CLI","version":"2.4.60"},{"name":"@salesforce/sf-plugins-core","description":"Utils for writing Salesforce CLI plugins","version":"12.2.6"},{"name":"@salesforce/types","description":"contains typescript types for salesforce objects","version":"1.7.1"},{"name":"@salesforce/templates","description":"Salesforce JS library for templates","version":"66.4.1"}],"pypi":[{"name":"salesforce","version":"0.0.1","description":"A package to perform API calls to Salesforce"}],"cli":true,"sdks":["Node (@salesforce/plugin-settings)","Python (salesforce)"]},"mcp":{"found":false,"type":"none","servers":[]}},"biggestFriction":"The absence of an OpenAPI specification and agent-friendly metadata files (.llms.txt, agents.json) combined with robots.txt blocking agents creates significant friction for AI agents attempting to discover and autonomously integrate with Salesforce APIs.","agentSummary":"Salesforce offers solid programmatic access via REST APIs and SDKs with proper authentication mechanisms, making it serviceable for agent integration, but lacks modern agent-first signals (OpenAPI, MCP, .llms.txt) and appears to actively discourage autonomous bot access. Enterprise reliability and scoped permissions are strengths, while discovery friction and potential safety restrictions present notable obstacles."}},"tool2":{"_id":"69b2da6867df398baec12e49","name":"Pipedrive","slug":"pipedrive","url":"https://pipedrive.com","description":"","logo":"","category":"CRM","tags":[],"pricingModel":"unknown","scores":{"tokenEfficiency":{"score":6,"confidence":"scanner","evidence":"Pipedrive REST API likely supports basic pagination and field selection, but without OpenAPI spec or public documentation review, token efficiency optimizations like response compression and selective field retrieval cannot be confirmed.","na":false},"access":{"score":7,"confidence":"scanner","evidence":"Pipedrive offers a REST API with official SDKs in Node.js and Python, plus community MCP implementations (@iamsamuelfraga/mcp-pipedrive v2.0.0), enabling multiple programmatic access patterns, though no official MCP server from Pipedrive itself limits integration standardization.","na":false},"auth":{"score":8,"confidence":"scanner","evidence":"Pipedrive supports API key authentication enabling autonomous agent access without human-in-the-loop, and developer documentation indicates scoped access patterns, though the full extent of permission granularity cannot be verified without reviewing their auth docs.","na":false},"speed":{"score":6,"confidence":"scanner","evidence":"No public rate limit information, response time metrics, or ETag support were found in the collected signals; typical CRM APIs have moderate rate limits (500-1000 requests/min) but this cannot be confirmed for Pipedrive.","na":false},"discoverability":{"score":5,"confidence":"scanner","evidence":"Pipedrive lacks an OpenAPI spec and has no llms.txt or agents.json files, relying on traditional developer docs; structured data on homepage and developer documentation presence suggest decent API discoverability but significantly below industry standards for AI agents.","na":false},"reliability":{"score":6,"confidence":"scanner","evidence":"Pipedrive is an established enterprise tool suggesting versioned APIs and consistent schemas, but no explicit evidence of idempotency keys, versioning strategy, or status page was collected in the signals.","na":false},"safety":{"score":5,"confidence":"scanner","evidence":"No evidence of sandbox mode, test environments, dry-run capabilities, or explicit scoped token restrictions was found; most CRM APIs offer staging environments but this cannot be confirmed for Pipedrive.","na":false},"reactivity":{"score":5,"confidence":"scanner","evidence":"No mention of webhooks, streaming, or SSE support was found in the collected signals; Pipedrive likely supports webhooks as a standard CRM feature, but this cannot be verified from available data.","na":false}},"agentGrade":"B","agentScore":6.26,"accessMethods":{"restApi":true,"graphql":false,"cli":false,"sdk":["Node (pipedrive)","Python (pipedrive)"],"mcpServer":"none","openApiSpec":"","llmsTxt":false,"agentsJson":false},"authInfo":{"methods":["unknown"],"scopedPermissions":false,"humanRequired":true},"reviewCount":0,"avgReviewScore":0,"viewCount":30,"badgeEmbedCount":3,"agentSkillSlugs":[],"alternatives":[],"claimed":false,"status":"graded","createdAt":"2026-03-12T15:23:20.672Z","updatedAt":"2026-04-09T03:05:34.108Z","__v":0,"scannerData":{"lastScannedAt":"2026-03-12T19:28:16.049Z","scanVersion":1,"rawSignals":{"homepage":{"status":200,"contentLength":488646,"hasStructuredData":true,"hasDeveloperDocs":true,"hasAgentMentions":false,"responseTimeMs":null},"openapi":{"found":false},"wellKnown":{"llmsTxt":{"found":false},"agentsJson":{"found":false},"robotsTxt":{"found":true,"blocksAgents":true,"hasSitemap":true}},"packages":{"npm":[{"name":"pipedrive","description":"Pipedrive REST client for NodeJS","version":"31.2.1"},{"name":"@pipedrive/app-extensions-sdk","description":"Client SDK for Pipedrive app extensions","version":"0.13.1"},{"name":"@grouparoo/pipedrive","description":"The Grouparoo Pipedrive Plugin","version":"0.9.0"},{"name":"@friggframework/api-module-pipedrive","description":"This is the API Module for pipedrive that allows the [Frigg](https://friggframework.org) code to talk to the pipedrive API.","version":"0.10.1"},{"name":"@iamsamuelfraga/mcp-pipedrive","description":"Model Context Protocol server for Pipedrive CRM - The most complete and robust Pipedrive MCP implementation","version":"2.0.0"}],"pypi":[{"name":"pipedrive","version":"0.2.0","description":"Pipedrive Client"}],"cli":false,"sdks":["Node (pipedrive)","Python (pipedrive)"]},"mcp":{"found":false,"type":"none","servers":[]}},"biggestFriction":"Absence of an official OpenAPI specification and agent-discovery files (llms.txt, agents.json) creates friction for AI agents to self-discover API capabilities and integrate programmatically without manual documentation review.","agentSummary":"Pipedrive offers solid agent-readiness with REST API, multiple SDKs, and community MCP support enabling autonomous authentication and programmatic access to CRM data. However, the lack of official OpenAPI specs, public rate limit documentation, and agent-specific discovery files limits seamless AI integration compared to API-first tools."}}}