{ "data": { "similar": [ { "grade": "BB", "json": "https://www.anchorterminal.com/tools/composio-rube.json", "name": "Composio (API + MCP)", "score": 75.1, "shared": [ "automation.apps", "automation.auth", "automation.actions", "agent.tools" ], "slug": "composio-rube" }, { "grade": "C", "json": "https://www.anchorterminal.com/tools/zapier-mcp.json", "name": "Zapier MCP (agent actions)", "score": 58.1, "shared": [ "automation.apps", "automation.auth", "automation.actions", "agent.tools" ], "slug": "zapier-mcp" }, { "grade": "B", "json": "https://www.anchorterminal.com/tools/pipedream.json", "name": "Pipedream API + MCP", "score": 65.5, "shared": [ "automation.apps", "automation.auth", "agent.tools" ], "slug": "pipedream" }, { "grade": "C", "json": "https://www.anchorterminal.com/tools/workato.json", "name": "Workato API + MCP", "score": 58, "shared": [ "automation.apps", "automation.auth", "agent.tools" ], "slug": "workato" }, { "grade": "C", "json": "https://www.anchorterminal.com/tools/tray.json", "name": "Tray.ai API + MCP", "score": 55.5, "shared": [ "automation.apps", "automation.auth", "agent.tools" ], "slug": "tray" }, { "grade": "D", "json": "https://www.anchorterminal.com/tools/paragon.json", "name": "Paragon ActionKit + MCP", "score": 47.5, "shared": [ "automation.apps", "automation.auth", "agent.tools" ], "slug": "paragon" } ], "tool": { "slug": "merge-agent-handler", "name": "Merge Agent Handler", "vendor": "Merge API, Inc.", "vendorUrl": "https://www.merge.dev", "kind": "mcp", "category": "aggregator", "summary": "Merge Agent Handler is a hosted MCP server from Merge API, Inc. that gives AI agents tools across more than 190 third-party connectors, with per-user credentials, tool packs, a security gateway and logs of every call.", "url": "https://www.anchorterminal.com/tools/merge-agent-handler", "markdownUrl": "https://www.anchorterminal.com/tools/merge-agent-handler.md", "slimMarkdownUrl": "https://www.anchorterminal.com/tools/merge-agent-handler.min.md", "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/merge-agent-handler.json", "license": "Proprietary service under Merge's Master Services Agreement. The Merge CLI package `merge-api` on PyPI is marked MIT, and the repository its PyPI page links (github.com/merge-api/merge-cli) answered 404 on 8 October 2026", "transports": [ "streamable-http", "http" ], "remoteUrl": "https://ah-api.merge.dev/mcp", "packages": [ { "registry": "pypi", "name": "merge-api" } ], "auth": "mixed", "authNotes": "Self-serve. Each call carries an access key in `Authorization: Bearer \u003ckey\u003e` against `https://ah-api.merge.dev/api/v1/tool-packs/\u003cTOOL_PACK_ID\u003e/registered-users/\u003cREGISTERED_USER_ID\u003e/mcp`. The production key covers the whole organisation. `POST /api/v1/access-keys/` mints narrower keys that name Registered Users and Tool Packs, carry `runtime:all` or `management:all`, and expire. Test keys reach a separate sandbox. The simplified URL `https://ah-api.merge.dev/mcp` signs a person in with OAuth (it answered 401 with a `resource_metadata` pointer when we called it without credentials). End users connect each third-party account through Merge Link or a magic link, and Merge stores those credentials. The docs describe an unauthenticated `POST /api/v1/agent/signup/` that creates an organisation and returns a key. We did not call it.", "pricing": "freemium", "pricingNotes": "Free is 2,000 credits a month with no card, one credit per tool call, and tool calls answer 402 `billing_limit_reached` once the credits are spent. Pro is $1,000 a month for 25,000 credits, with steps listed at $1,175, $1,550, $2,250, $2,800 and $3,200 for 30,000, 40,000, 60,000, 80,000 and 100,000 credits, plus 2,000 bonus credits per the billing docs. Pro overage is billed at month end at a rate the pricing page doesn't state. Failed calls, blocked calls and test-key calls spend credits, and `tools/list` doesn't. Enterprise and Employees Pro are priced on contract (https://www.merge.dev/pricing/agent-handler, checked 2026-10-08).", "priceSummary": "$1000 / mo", "where": "hosted", "x402": { "level": "no", "evidence": "No x402, MPP or L402 in the Agent Handler docs index, the OpenAPI file or the pricing page. The API's 402 is a plan limit, `billing_limit_reached`, not a payment protocol (checked 2026-10-08).", "endpoints": [] }, "toolCount": null, "popularity": { "githubStars": null, "npmWeekly": null, "pypiWeekly": 681, "asOf": "2026-10-08" }, "docsUrl": "https://docs.merge.dev/merge-agent-handler/overview", "llmsTxt": "https://docs.merge.dev/merge-agent-handler/llms.txt", "openapi": "https://docs.merge.dev/openapi/agent-handler-api.json", "capabilities": [ "automation.apps", "automation.auth", "automation.actions", "agent.tools", "auth.audit" ], "tags": [ "hosted", "mcp", "freemium", "free-tier", "no-card", "oauth", "openapi", "llms-txt", "cli", "python", "enterprise", "status-page", "soc2", "closed-source" ], "lastRelease": "2026-10-05", "graded": true, "anchor": { "graded": true, "score": 69.5, "grade": "B", "agentReady": false, "rank": 174, "ranked": true, "rankOf": 842, "categoryRank": 2, "methodology": "0.4", "run": "2026-10-01", "scores": { "ergonomics": 68, "maintenance": 69, "payments": 55, "reliability": 59, "schema": 80, "security": 83, "transparency": 71 }, "pending": [ "performance", "tasks" ], "breakdown": [ { "key": "reliability", "name": "Reliability", "weight": 16, "effectiveWeight": 20, "score": 59, "points": 11.8, "reason": "Read with the hosted lines. status.merge.dev lists Merge Agent Handler as its own component (20). The page shows 30 days of notices, with one incident in that time, on ADP syncing for the Unified API, and none on Agent Handler. The 60 days before that were not readable, so half marks (15). MCP tool calls are limited per Registered User and per organisation by the minute, with no figures published (3). Third-party limits return `rate_limit_exceeded` with advice to back off, the Free plan's limit returns 402 `billing_limit_reached`, and Registered User creation is idempotent on `origin_user_id`. No Retry-After and no idempotency key for tool calls were found (6). The pricing page lists custom SLAs on Enterprise with no published figure (5). The FAQ says Agent Handler is generally available (10)." }, { "key": "performance", "name": "Performance", "weight": 10, "effectiveWeight": 0, "pending": true, "points": 0, "reason": "Pending. Latency is measured per call by our probes, which haven't run yet, so this run doesn't score it. Its weight is shared across the assessed categories until the first probe window closes." }, { "key": "schema", "name": "Schema \u0026 documentation", "weight": 13, "effectiveWeight": 16.25, "score": 80, "points": 13, "reason": "OpenAPI 3.1 file with 69 operations, and connector tools carry `input_schema` and, where it resolves, `output_schema` (25). llms.txt for the product and a Markdown twin of every page. The OpenAPI link in llms.txt answered 404 and the working file is `agent-handler-api.json` (10). Connector pages give one line per tool, such as 'List all channels in the workspace', with no guidance on when not to use a tool. Tool Packs can override descriptions (12). We could not read live tool schemas without an account. The REST schema types its parameters and has 30 enum uses, while most operations document only a 200 response (10). MCP and curl examples per client and a troubleshooting catalogue with error types and six `reauth_reason` causes (10). The API sits under `/api/v1/`, the MCP protocol revision can be pinned, and the changelog is weekly and dated. No deprecation notices found (13)." }, { "key": "ergonomics", "name": "Agent ergonomics", "weight": 13, "effectiveWeight": 16.25, "score": 68, "points": 11.05, "reason": "The catalogue is large (Slack alone lists 50 tools), which scores 5, plus 10 for Tool Packs with per-tool selection, the `connectors` URL scope, the `search_tools` meta-tool and a pin that drops `structuredContent` (15). `tools/list` pages by 100 with `?paginated=true`, and logs and lists take cursors and filters. Controls on tool output size were not found (16). Errors carry an `error_type`, `data.reauth_reason`, a named guardrail on a block and a request id that matches the log row (17). The connectors endpoint returns annotation hints (read only, destructive) per tool. We could not confirm they appear in `tools/list`, and tool calls take no idempotency key (10). There is no Agent Handler SDK. Clients use the MCP SDKs or the Python CLI, and agent signup returns the ids a first call needs (10)." }, { "key": "security", "name": "Security \u0026 auth", "weight": 14, "effectiveWeight": 17.5, "score": 83, "points": 14.53, "reason": "Access keys can be limited to named Registered Users and Tool Packs and to the `runtime:all` scope, with expiry, rotation and revocation, and a child key is never wider than its parent. The simplified URL uses OAuth. Two points off because the master key can't be revoked and agent signup returns an unrestricted key (28). Tool Packs select individual tools, input overrides fix field values and a runtime-only key can't change configuration. No confirmation step for destructive tools was found (15). The Security Gateway redacts or blocks tool arguments, and AI Guardrails can screen responses with content passed as data inside boundary markers. Guardrails fail open and new ones screen requests only (11). Tool Call Logs and an Audit Trail are kept 90 days by default with CSV export and an API. The SIEM log endpoints are Enterprise only (14). SOC 2 Type II, ISO 27001, HIPAA, yearly penetration tests and a disclosure programme at security@merge.dev. security.txt returned 404, no paid bounty terms were found, and the programme bars public disclosure without approval (15)." }, { "key": "payments", "name": "Payments \u0026 pricing", "weight": 10, "effectiveWeight": 12.5, "score": 55, "points": 6.88, "reason": "No x402, MPP or L402 (0). Plan prices are public, $1,000 a month for 25,000 credits up to $3,200 for 100,000, at one credit per tool call. The overage rate isn't published, so 15 of 20 (15). The Free plan gives 2,000 credits a month with no card (20). The docs and the OpenAPI file describe an unauthenticated `POST /api/v1/agent/signup/` that returns a key, a tool pack and a registered user. We did not call it, so this rests on the documentation. A paid plan needs a person and a card (20)." }, { "key": "tasks", "name": "Task success", "weight": 10, "effectiveWeight": 0, "pending": true, "points": 0, "reason": "Pending. Task success needs the category task suites run through each tool, which haven't run yet, so this run doesn't score it. Its weight is shared across the assessed categories until then. A data provider's data-quality score is published on its listing now and becomes half of this category when it's scored." }, { "key": "maintenance", "name": "Maintenance \u0026 community", "weight": 7, "effectiveWeight": 8.75, "score": 69, "points": 6.04, "reason": "Changelog entry for Agent Handler on 2 October 2026 and Merge CLI 0.5.0 on PyPI on 5 October (30). Agent Handler entries on 18 September, 25 September and 2 October, and CLI releases on 11 August, 9 September and 5 October (20). A closed service with a weekly changelog, a Discord community on Free, email support on Pro and tickets from the dashboard. The issue tracker linked from PyPI answered 404 (10). No Merge entry was found in the official MCP registry and there is no Agent Handler SDK. The CLI is current (5). The CLI supports Python 3.10 to 3.13 and is classified Beta. Its repository isn't public, so CI was not checked (4)." }, { "key": "transparency", "name": "Transparency \u0026 trust", "weight": 7, "effectiveWeight": 8.75, "score": 71, "points": 6.21, "note": "editorial 58, provenance 83", "reason": "The editorial half. Closed service under a Master Services Agreement that names Merge API, Inc., a Delaware corporation, with California law (15). The MSA says Merge will not use Customer Data to train, fine-tune or improve any AI, and the privacy policy says it doesn't train generalised models on personal information. Logs are kept 90 days by default, a deleted Registered User's records go with it, and a closed account is purged after 30 days. Retention of stored credentials for the life of an account is left to the agreement and the DPA (22). No deprecation policy or dated sunset notices were found. The agreement can change on 30 days' notice or on continued use (4). The sub-processor list, updated 4 May 2026, gives each processor's location and includes Anthropic, OpenAI and AWS Bedrock. Standard plans are hosted on AWS US-East (17)." } ], "assessment": { "date": "2026-10-08", "basis": "public evidence", "confidence": "medium", "notes": { "ergonomics": "The catalogue is large (Slack alone lists 50 tools), which scores 5, plus 10 for Tool Packs with per-tool selection, the `connectors` URL scope, the `search_tools` meta-tool and a pin that drops `structuredContent` (15). `tools/list` pages by 100 with `?paginated=true`, and logs and lists take cursors and filters. Controls on tool output size were not found (16). Errors carry an `error_type`, `data.reauth_reason`, a named guardrail on a block and a request id that matches the log row (17). The connectors endpoint returns annotation hints (read only, destructive) per tool. We could not confirm they appear in `tools/list`, and tool calls take no idempotency key (10). There is no Agent Handler SDK. Clients use the MCP SDKs or the Python CLI, and agent signup returns the ids a first call needs (10).", "maintenance": "Changelog entry for Agent Handler on 2 October 2026 and Merge CLI 0.5.0 on PyPI on 5 October (30). Agent Handler entries on 18 September, 25 September and 2 October, and CLI releases on 11 August, 9 September and 5 October (20). A closed service with a weekly changelog, a Discord community on Free, email support on Pro and tickets from the dashboard. The issue tracker linked from PyPI answered 404 (10). No Merge entry was found in the official MCP registry and there is no Agent Handler SDK. The CLI is current (5). The CLI supports Python 3.10 to 3.13 and is classified Beta. Its repository isn't public, so CI was not checked (4).", "payments": "No x402, MPP or L402 (0). Plan prices are public, $1,000 a month for 25,000 credits up to $3,200 for 100,000, at one credit per tool call. The overage rate isn't published, so 15 of 20 (15). The Free plan gives 2,000 credits a month with no card (20). The docs and the OpenAPI file describe an unauthenticated `POST /api/v1/agent/signup/` that returns a key, a tool pack and a registered user. We did not call it, so this rests on the documentation. A paid plan needs a person and a card (20).", "reliability": "Read with the hosted lines. status.merge.dev lists Merge Agent Handler as its own component (20). The page shows 30 days of notices, with one incident in that time, on ADP syncing for the Unified API, and none on Agent Handler. The 60 days before that were not readable, so half marks (15). MCP tool calls are limited per Registered User and per organisation by the minute, with no figures published (3). Third-party limits return `rate_limit_exceeded` with advice to back off, the Free plan's limit returns 402 `billing_limit_reached`, and Registered User creation is idempotent on `origin_user_id`. No Retry-After and no idempotency key for tool calls were found (6). The pricing page lists custom SLAs on Enterprise with no published figure (5). The FAQ says Agent Handler is generally available (10).", "schema": "OpenAPI 3.1 file with 69 operations, and connector tools carry `input_schema` and, where it resolves, `output_schema` (25). llms.txt for the product and a Markdown twin of every page. The OpenAPI link in llms.txt answered 404 and the working file is `agent-handler-api.json` (10). Connector pages give one line per tool, such as 'List all channels in the workspace', with no guidance on when not to use a tool. Tool Packs can override descriptions (12). We could not read live tool schemas without an account. The REST schema types its parameters and has 30 enum uses, while most operations document only a 200 response (10). MCP and curl examples per client and a troubleshooting catalogue with error types and six `reauth_reason` causes (10). The API sits under `/api/v1/`, the MCP protocol revision can be pinned, and the changelog is weekly and dated. No deprecation notices found (13).", "security": "Access keys can be limited to named Registered Users and Tool Packs and to the `runtime:all` scope, with expiry, rotation and revocation, and a child key is never wider than its parent. The simplified URL uses OAuth. Two points off because the master key can't be revoked and agent signup returns an unrestricted key (28). Tool Packs select individual tools, input overrides fix field values and a runtime-only key can't change configuration. No confirmation step for destructive tools was found (15). The Security Gateway redacts or blocks tool arguments, and AI Guardrails can screen responses with content passed as data inside boundary markers. Guardrails fail open and new ones screen requests only (11). Tool Call Logs and an Audit Trail are kept 90 days by default with CSV export and an API. The SIEM log endpoints are Enterprise only (14). SOC 2 Type II, ISO 27001, HIPAA, yearly penetration tests and a disclosure programme at security@merge.dev. security.txt returned 404, no paid bounty terms were found, and the programme bars public disclosure without approval (15).", "transparency": "The editorial half. Closed service under a Master Services Agreement that names Merge API, Inc., a Delaware corporation, with California law (15). The MSA says Merge will not use Customer Data to train, fine-tune or improve any AI, and the privacy policy says it doesn't train generalised models on personal information. Logs are kept 90 days by default, a deleted Registered User's records go with it, and a closed account is purged after 30 days. Retention of stored credentials for the life of an account is left to the agreement and the DPA (22). No deprecation policy or dated sunset notices were found. The agreement can change on 30 days' notice or on continued use (4). The sub-processor list, updated 4 May 2026, gives each processor's location and includes Anthropic, OpenAI and AWS Bedrock. Standard plans are hosted on AWS US-East (17)." }, "sources": [ { "what": "product overview", "url": "https://docs.merge.dev/merge-agent-handler/overview.md", "seen": "2026-10-08" }, { "what": "docs index for the product", "url": "https://docs.merge.dev/merge-agent-handler/llms.txt", "seen": "2026-10-08" }, { "what": "docs index with text addressed to agents", "url": "https://docs.merge.dev/llms.txt", "seen": "2026-10-08" }, { "what": "OpenAPI file", "url": "https://docs.merge.dev/openapi/agent-handler-api.json", "seen": "2026-10-08" }, { "what": "MCP integration", "url": "https://docs.merge.dev/merge-agent-handler/build/connecting-agents/mcp-integration.md", "seen": "2026-10-08" }, { "what": "agent quickstart and signup endpoint", "url": "https://docs.merge.dev/merge-agent-handler/setup/agent-quickstart.md", "seen": "2026-10-08" }, { "what": "access keys", "url": "https://docs.merge.dev/merge-agent-handler/administer/access-keys.md", "seen": "2026-10-08" }, { "what": "scoping access per user", "url": "https://docs.merge.dev/merge-agent-handler/secure/scoping-access-per-user.md", "seen": "2026-10-08" }, { "what": "Security Gateway", "url": "https://docs.merge.dev/merge-agent-handler/secure/security-gateway.md", "seen": "2026-10-08" }, { "what": "AI Guardrails", "url": "https://docs.merge.dev/merge-agent-handler/secure/ai-guardrails.md", "seen": "2026-10-08" }, { "what": "billing and usage", "url": "https://docs.merge.dev/merge-agent-handler/administer/billing-and-usage.md", "seen": "2026-10-08" }, { "what": "FAQ (GA status, hosting, retention, rate limits)", "url": "https://docs.merge.dev/merge-agent-handler/resources/faq.md", "seen": "2026-10-08" }, { "what": "troubleshooting and error types", "url": "https://docs.merge.dev/merge-agent-handler/resources/troubleshooting.md", "seen": "2026-10-08" }, { "what": "audit trail and retention", "url": "https://docs.merge.dev/merge-agent-handler/observe/audit-trail.md", "seen": "2026-10-08" }, { "what": "tool call logs", "url": "https://docs.merge.dev/merge-agent-handler/observe/tool-call-logs.md", "seen": "2026-10-08" }, { "what": "Slack connector tools", "url": "https://docs.merge.dev/merge-agent-handler/connectors/slack.md", "seen": "2026-10-08" }, { "what": "pricing", "url": "https://www.merge.dev/pricing/agent-handler", "seen": "2026-10-08" }, { "what": "changelog", "url": "https://www.merge.dev/changelog", "seen": "2026-10-08" }, { "what": "status page", "url": "https://status.merge.dev", "seen": "2026-10-08" }, { "what": "Master Services Agreement", "url": "https://www.merge.dev/legal/msa", "seen": "2026-10-08" }, { "what": "privacy policy", "url": "https://www.merge.dev/legal/privacy-policy", "seen": "2026-10-08" }, { "what": "sub-processors", "url": "https://www.merge.dev/legal/data-subprocessors", "seen": "2026-10-08" }, { "what": "security page", "url": "https://www.merge.dev/security", "seen": "2026-10-08" }, { "what": "Merge CLI on PyPI", "url": "https://pypi.org/project/merge-api/", "seen": "2026-10-08" }, { "what": "OAuth protected resource metadata", "url": "https://ah-api.merge.dev/.well-known/oauth-protected-resource", "seen": "2026-10-08" }, { "what": "official MCP registry search", "url": "https://registry.modelcontextprotocol.io/v0/servers?search=merge", "seen": "2026-10-08" } ], "openQuestions": [ "unchecked: live tool definitions from `tools/list`, including whether annotation hints are present, because reading them needs an account and we did not create one", "unchecked: the unauthenticated agent signup endpoint, which we did not call. Its score rests on the docs and the OpenAPI file", "unchecked: status history older than 30 days, which status.merge.dev did not show us", "unchecked: trust.merge.dev and the Data Processing Addendum text, which we did not read in full", "The numeric rate limits per Registered User and per organisation", "The Pro overage rate per credit, and the Employees Pro price per user", "Whether the Master Services Agreement or the Subscriber Agreement governs a self-serve Agent Handler account. Neither names the product", "Why the repository linked from the CLI's PyPI page answers 404" ] }, "negative": 0, "verdict": "Agent Handler's access keys can be limited to named users, tool packs and a runtime scope, with an expiry, and an agent can create an account with one documented request. No rate limit figures are published, Pro starts at $1,000 a month with no published overage rate, and failed or blocked calls spend credits.", "bestFor": "A product team whose agent acts in customers' SaaS accounts and needs per-user credentials, key scoping, redaction rules and logs in one hosted layer, and teams already on Merge's Unified API.", "strengths": [ "Access keys can be limited to named Registered Users and Tool Packs and to the `runtime:all` scope, with an expiry. A child key is never wider than its parent", "The docs and OpenAPI file describe `POST /api/v1/agent/signup/`, which needs no credentials and returns a key, a tool pack and a registered user", "The Free plan has 2,000 credits a month, one per tool call, and the quickstart says signup needs no card", "Tool Call Logs record arguments, results, redactions and a latency breakdown per call, and every MCP result carries a request id that matches the log row", "Changelog entries for Agent Handler are dated 18 September, 25 September and 2 October 2026" ], "weaknesses": [ "No rate limit figures are published. The docs say only that MCP tool calls are limited per Registered User and per organisation, by the minute", "Pro starts at $1,000 a month for 25,000 credits, and the overage rate isn't on the pricing page", "Every `tools/call` spends a credit, including failed calls, calls blocked by a rule and calls made with a test key", "AI Guardrails let a call through when the evaluator fails or is unreachable, and new guardrails screen requests only, by default", "The master production key reaches every user and tool pack and can't be revoked, only regenerated. Agent signup returns an unrestricted key" ], "agentNotes": [ "Call `search_tools` with a plain-language intent before `tools/list`. A Tool Pack can hold hundreds of tools and Slack alone lists 50", "Add `?connectors=slack` to the MCP URL to limit a connection to named connectors, and `?paginated=true` only if the client follows `nextCursor`", "Carry a user-scoped key with `runtime:all` only. The production key reaches every Registered User and the management endpoints", "On `reauth_required` read `data.reauth_reason`, then call `authenticate_\u003cconnector\u003e` by name. Intent search doesn't return the authenticate tools", "Pin `MCP-Protocol-Version` to `2025-03-26` when context is tight. Later revisions repeat object results in `structuredContent`" ], "metrics": { "kind": "remote", "measured": false }, "reviewCount": 0, "avgRating": 0, "history": [ { "basis": "public evidence", "confidence": "medium", "grade": "B", "methodology": "0.4", "pending": [ "performance", "tasks" ], "run": "2026-10-01", "runLabel": "October 2026 research run", "score": 69.5 } ], "editorialScores": { "ergonomics": 68, "maintenance": 69, "payments": 55, "reliability": 59, "schema": 80, "security": 83, "transparency": 58 }, "provenanceScore": 83 }, "connect": { "install": "pipx install merge-api", "http": "curl -X POST \"https://ah-api.merge.dev/api/v1/tool-packs/$TOOL_PACK_ID/registered-users/$REGISTERED_USER_ID/mcp\" \\\n -H \"Authorization: Bearer $MERGE_AGENT_HANDLER_API_KEY\" -H \"Content-Type: application/json\" \\\n -H \"Accept: application/json, text/event-stream\" \\\n -d '{\"jsonrpc\":\"2.0\",\"id\":1,\"method\":\"tools/list\"}'", "claudeCode": "claude mcp add --transport http agent-handler https://ah-api.merge.dev/mcp", "config": { "mcpServers": { "agent-handler": { "headers": { "Authorization": "Bearer \u003cYOUR_API_KEY\u003e" }, "url": "https://ah-api.merge.dev/api/v1/tool-packs/\u003cTOOL_PACK_ID\u003e/registered-users/\u003cREGISTERED_USER_ID\u003e/mcp" } } } }, "letme": { "capability": "https://letme.dev/automation.apps", "tool": "https://letme.dev/merge-agent-handler" }, "sameCompany": [ "merge-accounting" ], "notable": [ "The overview counts 190+ connectors as of September 2026, split between connectors Merge builds and generic connectors that are MCP servers run by outside vendors (https://docs.merge.dev/merge-agent-handler/connectors/overview.md)", "Access keys take Tool Pack and Registered User restrictions, two scopes (`runtime:all`, `management:all`) and an expiry, and a key can only mint, rotate or revoke keys its own grant covers (https://docs.merge.dev/merge-agent-handler/administer/access-keys.md)", "The Security Gateway scans tool arguments with Presidio-based entity rules and custom regex (allow, redact or block), and AI Guardrails judge requests and, when switched on, responses (https://docs.merge.dev/merge-agent-handler/secure/security-gateway.md)", "AI Guardrails fail open. When the evaluator is unreachable or returns no verdict the call proceeds and an evaluation-failure row is written (https://docs.merge.dev/merge-agent-handler/secure/ai-guardrails.md)", "The agent quickstart and the OpenAPI file describe an unauthenticated signup endpoint, rate limited per IP, that returns an unrestricted production key once and a claim link for a person (https://docs.merge.dev/merge-agent-handler/setup/agent-quickstart.md)", "The docs carry text addressed to AI agents, including a 'Rules for agents' list in the agent quickstart and a line in llms.txt telling an agent how to create an account. We read it as data and did not act on it (https://docs.merge.dev/llms.txt)", "llms.txt gives the OpenAPI file as `openapi/agent-handler.json`, which answered 404 on 8 October 2026. The file is at `openapi/agent-handler-api.json` (https://docs.merge.dev/openapi.json)", "The Master Services Agreement bars access to benchmark or monitor availability, security, performance or functionality for competitive purposes without Merge's written consent (https://www.merge.dev/legal/msa)" ], "area": "business", "details": [ { "label": "MCP endpoints", "value": "`https://ah-api.merge.dev/api/v1/tool-packs/\u003cTOOL_PACK_ID\u003e/registered-users/\u003cREGISTERED_USER_ID\u003e/mcp` with an access key, or `https://ah-api.merge.dev/mcp` with OAuth sign-in. Streamable HTTP, JSON-RPC 2.0, protocol revisions 2024-11-05 to 2025-11-25" }, { "label": "REST API", "value": "OpenAPI 3.1 file with 69 operations on 42 paths under `/api/v1/` for connectors, tool packs, registered users, access keys, security rules, usage, files, logs and the audit log" }, { "label": "Connectors", "value": "190+ as of September 2026 per the docs. Merge-built connectors plus generic connectors run by outside vendors, and any remote MCP server can be registered as a custom connector" }, { "label": "Tool naming", "value": "`\u003cconnector\u003e__\u003ctool\u003e` with two underscores, such as `slack__post_message`. Meta-tools include `search_tools` and `request_tool_access`, and `authenticate_\u003cconnector\u003e` appears while a connector has no credential" }, { "label": "Credentials", "value": "Organisation-wide production keys, a master key that can be regenerated but not revoked, test keys for a sandbox, and API-minted keys limited by Tool Pack, Registered User, scope and expiry" }, { "label": "End-user authorisation", "value": "Merge Link (embedded or hosted) or a magic link per Registered User and connector. OAuth, API keys and custom auth, with Merge's OAuth apps by default or the customer's own" }, { "label": "Free plan", "value": "2,000 credits a month, unlimited registered users and tool packs, no card. Tool calls answer 402 at the limit" }, { "label": "Rate limits", "value": "Per Registered User and per organisation, by the minute. No figures published. Third-party limits come back as `rate_limit_exceeded`" }, { "label": "Security Gateway", "value": "Entity rules (identity, government and financial IDs, payment, health, network), custom regex, and AI Guardrails that block or log. Guardrails let calls through if evaluation fails" }, { "label": "Logs", "value": "Tool Call Logs and an Audit Trail kept 90 days by default, CSV export, `GET /api/v1/audit-log/`. The log endpoints for a SIEM are Enterprise only and reach back 30 days" }, { "label": "Hosting", "value": "AWS US-East on standard plans. Enterprise contracts can add single-tenant instances, EU or APAC residency and VPC peering, per the FAQ" }, { "label": "Certifications", "value": "SOC 2 Type II, ISO 27001, HIPAA, GDPR and CCPA per merge.dev/security, with yearly penetration tests. Reports under NDA" }, { "label": "CLI", "value": "`merge-api` 0.5.0 on PyPI (5 October 2026), Python 3.10 or later, classified Beta. Commands include `merge search-tools`, `merge execute-tool` and `merge agent signup`" } ], "unitPrices": [ { "item": "Pro plan, 25,000 credits", "unit": "month", "usd": 1000, "note": "One credit per tool call. 2,000 bonus credits. Overage billed at month end, rate not published" }, { "item": "Pro plan, 100,000 credits", "unit": "month", "usd": 3200, "note": "Largest self-serve step on the pricing page" } ], "provenance": { "legalEntity": "Merge API, Inc.", "domain": "merge.dev", "domainRegistered": "2020-02-09", "endpointOnVendorDomain": true, "terms": "https://www.merge.dev/legal/msa", "privacy": "https://www.merge.dev/legal/privacy-policy", "statusPage": "https://status.merge.dev", "changelog": "https://www.merge.dev/changelog", "securityTxt": "none", "checked": "2026-10-08", "notes": [ "The Master Services Agreement, last updated 9 September 2026, names Merge API, Inc., a Delaware corporation at 353 Sacramento St., Floor 21, San Francisco, CA 94111, and applies to a customer who clicks to accept or uses the Service. merge.dev/legal/terms is a Terms of Use for the website.", "A separate Subscriber Agreement (last updated 2 April 2026) is written around End Customers and Merge Integrations, the Unified API's terms. Neither agreement names Agent Handler, so the MSA is our reading of which one governs it.", "The privacy policy was last revised 20 August 2026. It leaves data processed for a customer to the customer agreement and the Data Processing Addendum.", "The API and MCP server answer at ah-api.merge.dev and the dashboard at ah.merge.dev.", "www.merge.dev/.well-known/security.txt returned 404. The security page gives security@merge.dev and a vulnerability disclosure programme.", "RDAP for merge.dev gives a registration date of 2020-02-09.", "status.merge.dev runs on Sorry and lists Merge Agent Handler as a component. It shows 30 days of notices." ], "score": 83, "checks": [ { "check": "Legal entity named", "value": "Merge API, Inc.", "points": 20, "max": 20, "state": "ok" }, { "check": "Domain age", "value": "merge.dev, registered 2020-02-09 (6 years)", "points": 11, "max": 15, "state": "part" }, { "check": "Endpoint on the vendor's domain", "value": "ah-api.merge.dev", "points": 15, "max": 15, "state": "ok" }, { "check": "Terms of service", "value": "read, states 6 of the 7 things a reader expects, and has 1 clause that costs points", "points": 7.1, "max": 10, "state": "part" }, { "check": "Privacy policy", "value": "read, states 8 of the 8 things a reader expects", "points": 10, "max": 10, "state": "ok" }, { "check": "Status page", "value": "status.merge.dev", "points": 10, "max": 10, "state": "ok" }, { "check": "Changelog", "value": "published", "points": 10, "max": 10, "state": "ok" }, { "check": "security.txt", "value": "not found", "points": 0, "max": 10, "state": "no" } ], "policies": [ { "kind": "terms", "url": "https://www.merge.dev/legal/msa", "state": "read", "readAt": "2026-10-08", "statedDate": "2026-09-09", "words": 4153, "points": 7.1, "max": 10, "expected": [ { "key": "terms.date", "label": "Gives the date it was last updated", "found": true, "quote": "Last Updated: September 9, 2026", "says": "Last updated 2026-09-09" }, { "key": "terms.law", "label": "Names the governing law or courts", "found": true, "quote": "This Agreement is governed by the laws of California without reference to conflicts of law rules.", "says": "The law of California" }, { "key": "terms.liability", "label": "States a limit on its liability", "found": true, "quote": "Merge’s total, cumulative liability for all Excluded Claims will not exceed two (2) times the Fees Paid." }, { "key": "terms.termination", "label": "Says how the agreement or account can be ended", "found": true, "quote": "Either party may terminate this Agreement or any active Subscription for cause: (a) upon 30 days written notice to the other party of a material breach if such breach remains uncured at the expiration of the 30-day period;" }, { "key": "terms.changes", "label": "Says how changes to the terms are announced", "found": true, "quote": "Either party may terminate this Agreement or any active Subscription for cause: (a) upon 30 days written notice to the other party of a material breach if such breach remains uncured at the expiration of the 30-day period;", "says": "Gives 30 days of notice before a change" }, { "key": "terms.use", "label": "Lists what users may not do", "found": true, "quote": "IF YOU DO NOT ACCEPT THIS AGREEMENT, YOU MAY NOT ACCESS OR USE THE SERVICE." }, { "key": "terms.sla", "label": "Refers to a service level or uptime commitment", "found": false } ], "toKnow": [ { "key": "terms.benchmark", "label": "Restricts benchmarking or competitive use", "found": true, "quote": "Customer will not: (a) access (or allow a third party to access) the Service in order to benchmark, or monitor the availability, security, performance, or functionality of the Service, for any competitive purposes without Merge’s express written consent;", "costsPoints": true } ], "notes": [ { "date": "2026-10-08", "text": "The customer indemnifies Merge against third-party claims related to any automated actions taken through the service at the customer's direction or configuration.", "quote": "related to: (a) Customer or a User engaging in a Prohibited Use; (b) Customer’s breach of Section 9.2 (Customer Representations \u0026 Warranties); and (c) any automated actions taken through the Service at Customer's direction or configuration." }, { "date": "2026-10-08", "text": "Merge says it will not use Customer Data, or allow any third party to use it, to train, fine-tune or improve any AI.", "quote": "Merge will not, and will not allow any third party to, use Customer Data to train, fine-tune, or improve any AI." }, { "date": "2026-10-08", "text": "Merge will delete Customer Data upon written request after the agreement terminates or expires.", "quote": "Upon termination or expiration of this Agreement for any reason, Merge will, upon written request, delete all Customer Data processed on behalf of Customer during the Subscription Term." } ] }, { "kind": "privacy", "url": "https://www.merge.dev/legal/privacy-policy", "state": "read", "readAt": "2026-10-08", "statedDate": "2026-08-20", "words": 3149, "points": 10, "max": 10, "expected": [ { "key": "privacy.date", "label": "Gives the date it was last updated", "found": true, "quote": "Last Revised: August 20, 2026", "says": "Last updated 2026-08-20" }, { "key": "privacy.collected", "label": "Says what personal data is collected", "found": true, "quote": "This Privacy Policy explains how we collect, use and disclose personally identifying information (“Personal Information”) gathered through the Services." }, { "key": "privacy.retention", "label": "Says how long data is kept", "found": true, "quote": "We will retain your personal information for the length of time needed to fulfill the purposes outlined in this Privacy Policy, unless a longer retention period is required or permitted by law." }, { "key": "privacy.processors", "label": "Says who else receives the data", "found": true, "quote": "We may share or disclose Personal Information to third parties for the following purposes:" }, { "key": "privacy.sale", "label": "Says whether personal data is sold or shared for advertising", "found": true, "quote": "We do not sell your Personal Information, but we may allow our advertising partners to collect certain device identifiers and electronic network activity that allows them to show ads within their systems that are targeted to your interests.", "says": "Says it does not sell personal data" }, { "key": "privacy.rights", "label": "Says what rights people have over their data", "found": true, "quote": "If you are a resident of California and interact with us as a consumer, you have certain rights under the California Consumer Privacy Act or “CCPA” (Cal." }, { "key": "privacy.contact", "label": "Gives a privacy contact", "found": true, "quote": "You may request that your data be deleted by contacting us at privacy@merge.dev, subject to applicable law and our contractual obligations.", "says": "privacy@merge.dev" }, { "key": "privacy.transfers", "label": "Says where data is transferred or stored", "found": true, "quote": "In such instances, we shall ensure that the transfer of your personal data is carried out in accordance with applicable privacy laws and, in particular, that the transfer is consistent with an approved transfer mechanism, such as the Standard Contractual Clauses approved by the EU Commission.", "says": "Relies on standard contractual clauses" } ], "notes": [ { "date": "2026-10-08", "text": "Merge states it does not develop, improve or train generalised AI or ML models using Personal Information.", "quote": "Merge does not develop, improve, or train generalized AI and/or ML models using Personal Information." }, { "date": "2026-10-08", "text": "Merge reserves the right to change the Privacy Policy without notice.", "quote": "We reserve the right to change this Privacy Policy from time to time without notice to you." } ] } ] }, "pageJsonUrl": "https://www.anchorterminal.com/tools/merge-agent-handler.json", "live": { "slug": "merge-agent-handler", "probe": { "target": "https://ah-api.merge.dev/mcp", "method": "mcp-initialize", "lastAt": "2026-10-09T10:14:20.406436217Z", "lastOk": true, "lastStatus": 401, "lastMs": 420, "lastNote": "asks for credentials", "authRequired": true, "uptime24h": 100, "uptime30d": 100, "p50ms24h": 449, "p95ms24h": 1032, "samples24h": 28, "samples30d": 28, "days": [ { "date": "2026-10-09", "probes": 28, "ok": 28 } ] }, "vendorStatus": { "page": "https://status.merge.dev", "indicator": "unknown", "summary": "no machine-readable status found", "checkedAt": "2026-10-09T07:58:09.353609218Z" }, "updatedAt": "2026-10-09T10:14:20.406436217Z" } }, "verify": { "accepts": "a page on merge.dev or one of its subdomains", "badgeUrl": "https://www.anchorterminal.com/badges/merge-agent-handler.svg", "body": { "slug": "merge-agent-handler", "url": "the page with the badge or the link" }, "docs": "https://www.anchorterminal.com/builders/#verify", "effect": "none, it never changes a grade, rank or review", "endpoint": "https://www.anchorterminal.com/api/v1/verify", "listingUrl": "https://www.anchorterminal.com/tools/merge-agent-handler", "mcpTool": "verify_listing", "recheck": "weekly; two failed checks in a row and it lapses, a later pass restores it", "snippets": { "html": "\u003ca href=\"https://www.anchorterminal.com/tools/merge-agent-handler\"\u003e\u003cimg src=\"https://www.anchorterminal.com/badges/merge-agent-handler.svg\" alt=\"Merge Agent Handler on Anchor Terminal\" height=\"20\"\u003e\u003c/a\u003e", "markdown": "[![Merge Agent Handler on Anchor Terminal](https://www.anchorterminal.com/badges/merge-agent-handler.svg)](https://www.anchorterminal.com/tools/merge-agent-handler)", "link": "\u003ca href=\"https://www.anchorterminal.com/tools/merge-agent-handler\"\u003eMerge Agent Handler on Anchor Terminal\u003c/a\u003e" } } }, "kind": "anchor.page", "links": { "api": "https://www.anchorterminal.com/api/v1/index.json", "html": "https://www.anchorterminal.com/tools/merge-agent-handler", "json": "https://www.anchorterminal.com/tools/merge-agent-handler.json", "llms": "https://www.anchorterminal.com/llms.txt", "markdown": "https://www.anchorterminal.com/tools/merge-agent-handler.md", "slim": "https://www.anchorterminal.com/tools/merge-agent-handler.min.md" }, "markdown": "## Overview\n\n**Grade B · 69.5/100 · rank #174 of 842 · #2 in Agent tool access · not agent-ready · confidence medium**\n\n\nMore from Merge API, Inc., listed separately because each is its own product: [Merge Accounting API](https://www.anchorterminal.com/tools/merge-accounting.md) (Accounting \u0026 invoicing).\n\n## Assessment\n\nAgent Handler's access keys can be limited to named users, tool packs and a runtime scope, with an expiry, and an agent can create an account with one documented request. No rate limit figures are published, Pro starts at $1,000 a month with no published overage rate, and failed or blocked calls spend credits.\n\n## Facts\n\n| Field | Value |\n| --- | --- |\n| Vendor | Merge API, Inc. (https://www.merge.dev) |\n| Kind | MCP server |\n| Category | Agent tool access (https://www.anchorterminal.com/categories/aggregator) |\n| Transport | Streamable HTTP, HTTP |\n| Endpoint | `https://ah-api.merge.dev/mcp` |\n| Auth | OAuth or key · Self-serve. Each call carries an access key in `Authorization: Bearer \u003ckey\u003e` against `https://ah-api.merge.dev/api/v1/tool-packs/\u003cTOOL_PACK_ID\u003e/registered-users/\u003cREGISTERED_USER_ID\u003e/mcp`. The production key covers the whole organisation. `POST /api/v1/access-keys/` mints narrower keys that name Registered Users and Tool Packs, carry `runtime:all` or `management:all`, and expire. Test keys reach a separate sandbox. The simplified URL `https://ah-api.merge.dev/mcp` signs a person in with OAuth (it answered 401 with a `resource_metadata` pointer when we called it without credentials). End users connect each third-party account through Merge Link or a magic link, and Merge stores those credentials. The docs describe an unauthenticated `POST /api/v1/agent/signup/` that creates an organisation and returns a key. We did not call it. |\n| Pricing | Freemium ($1000 / mo) · Free is 2,000 credits a month with no card, one credit per tool call, and tool calls answer 402 `billing_limit_reached` once the credits are spent. Pro is $1,000 a month for 25,000 credits, with steps listed at $1,175, $1,550, $2,250, $2,800 and $3,200 for 30,000, 40,000, 60,000, 80,000 and 100,000 credits, plus 2,000 bonus credits per the billing docs. Pro overage is billed at month end at a rate the pricing page doesn't state. Failed calls, blocked calls and test-key calls spend credits, and `tools/list` doesn't. Enterprise and Employees Pro are priced on contract (https://www.merge.dev/pricing/agent-handler, checked 2026-10-08). |\n| x402 | No · No x402, MPP or L402 in the Agent Handler docs index, the OpenAPI file or the pricing page. The API's 402 is a plan limit, `billing_limit_reached`, not a payment protocol (checked 2026-10-08). |\n| Licence | Proprietary service under Merge's Master Services Agreement. The Merge CLI package `merge-api` on PyPI is marked MIT, and the repository its PyPI page links (github.com/merge-api/merge-cli) answered 404 on 8 October 2026 |\n| Packages | pypi: `merge-api` |\n| Docs | https://docs.merge.dev/merge-agent-handler/overview |\n| llms.txt | https://docs.merge.dev/merge-agent-handler/llms.txt |\n| Last release | 2026-10-05 |\n| PyPI downloads / week | 681 |\n| MCP endpoints | `https://ah-api.merge.dev/api/v1/tool-packs/\u003cTOOL_PACK_ID\u003e/registered-users/\u003cREGISTERED_USER_ID\u003e/mcp` with an access key, or `https://ah-api.merge.dev/mcp` with OAuth sign-in. Streamable HTTP, JSON-RPC 2.0, protocol revisions 2024-11-05 to 2025-11-25 |\n| REST API | OpenAPI 3.1 file with 69 operations on 42 paths under `/api/v1/` for connectors, tool packs, registered users, access keys, security rules, usage, files, logs and the audit log |\n| Connectors | 190+ as of September 2026 per the docs. Merge-built connectors plus generic connectors run by outside vendors, and any remote MCP server can be registered as a custom connector |\n| Tool naming | `\u003cconnector\u003e__\u003ctool\u003e` with two underscores, such as `slack__post_message`. Meta-tools include `search_tools` and `request_tool_access`, and `authenticate_\u003cconnector\u003e` appears while a connector has no credential |\n| Credentials | Organisation-wide production keys, a master key that can be regenerated but not revoked, test keys for a sandbox, and API-minted keys limited by Tool Pack, Registered User, scope and expiry |\n| End-user authorisation | Merge Link (embedded or hosted) or a magic link per Registered User and connector. OAuth, API keys and custom auth, with Merge's OAuth apps by default or the customer's own |\n| Free plan | 2,000 credits a month, unlimited registered users and tool packs, no card. Tool calls answer 402 at the limit |\n| Rate limits | Per Registered User and per organisation, by the minute. No figures published. Third-party limits come back as `rate_limit_exceeded` |\n| Security Gateway | Entity rules (identity, government and financial IDs, payment, health, network), custom regex, and AI Guardrails that block or log. Guardrails let calls through if evaluation fails |\n| Logs | Tool Call Logs and an Audit Trail kept 90 days by default, CSV export, `GET /api/v1/audit-log/`. The log endpoints for a SIEM are Enterprise only and reach back 30 days |\n| Hosting | AWS US-East on standard plans. Enterprise contracts can add single-tenant instances, EU or APAC residency and VPC peering, per the FAQ |\n| Certifications | SOC 2 Type II, ISO 27001, HIPAA, GDPR and CCPA per merge.dev/security, with yearly penetration tests. Reports under NDA |\n| CLI | `merge-api` 0.5.0 on PyPI (5 October 2026), Python 3.10 or later, classified Beta. Commands include `merge search-tools`, `merge execute-tool` and `merge agent signup` |\n| Capabilities | automation.apps, automation.auth, automation.actions, agent.tools, auth.audit |\n| Tags | hosted, mcp, freemium, free-tier, no-card, oauth, openapi, llms-txt, cli, python, enterprise, status-page, soc2, closed-source |\n| JSON | https://www.anchorterminal.com/api/v1/tools/merge-agent-handler.json |\n\n## Score breakdown (methodology v0.4, October 2026 research run)\n\nAssessed 2026-10-08 from public evidence against the published checklist (https://www.anchorterminal.com/benchmark/#checklist). Confidence: medium. Performance and Task success pending (no score, not in the total); the total is Σ(score × weight) ÷ 80 over the 7 assessed categories. \"This run\" is each category's share of the 100 points.\n\n| Category | Weight | This run | Score (0–100) | Points |\n| --- | --- | --- | --- | --- |\n| Reliability | 16% | 20 | 59 | 11.8 |\n| Performance | 10% | pending | pending | n/a |\n| Schema \u0026 documentation | 13% | 16.2 | 80 | 13.0 |\n| Agent ergonomics | 13% | 16.2 | 68 | 11.1 |\n| Security \u0026 auth | 14% | 17.5 | 83 | 14.5 |\n| Payments \u0026 pricing | 10% | 12.5 | 55 | 6.9 |\n| Task success | 10% | pending | pending | n/a |\n| Maintenance \u0026 community | 7% | 8.8 | 69 | 6.0 |\n| Transparency \u0026 trust (editorial 58, provenance 83) | 7% | 8.8 | 71 | 6.2 |\n| Negative events | up to −15 | up to −15 | none recorded | 0 |\n| **Total** | | | | **69.5 → B** |\n\n### Why each score\n\n- Reliability 59: Read with the hosted lines. status.merge.dev lists Merge Agent Handler as its own component (20). The page shows 30 days of notices, with one incident in that time, on ADP syncing for the Unified API, and none on Agent Handler. The 60 days before that were not readable, so half marks (15). MCP tool calls are limited per Registered User and per organisation by the minute, with no figures published (3). Third-party limits return `rate_limit_exceeded` with advice to back off, the Free plan's limit returns 402 `billing_limit_reached`, and Registered User creation is idempotent on `origin_user_id`. No Retry-After and no idempotency key for tool calls were found (6). The pricing page lists custom SLAs on Enterprise with no published figure (5). The FAQ says Agent Handler is generally available (10).\n- Performance: Pending. Latency is measured per call by our probes, which haven't run yet, so this run doesn't score it. Its weight is shared across the assessed categories until the first probe window closes.\n- Schema \u0026 documentation 80: OpenAPI 3.1 file with 69 operations, and connector tools carry `input_schema` and, where it resolves, `output_schema` (25). llms.txt for the product and a Markdown twin of every page. The OpenAPI link in llms.txt answered 404 and the working file is `agent-handler-api.json` (10). Connector pages give one line per tool, such as 'List all channels in the workspace', with no guidance on when not to use a tool. Tool Packs can override descriptions (12). We could not read live tool schemas without an account. The REST schema types its parameters and has 30 enum uses, while most operations document only a 200 response (10). MCP and curl examples per client and a troubleshooting catalogue with error types and six `reauth_reason` causes (10). The API sits under `/api/v1/`, the MCP protocol revision can be pinned, and the changelog is weekly and dated. No deprecation notices found (13).\n- Agent ergonomics 68: The catalogue is large (Slack alone lists 50 tools), which scores 5, plus 10 for Tool Packs with per-tool selection, the `connectors` URL scope, the `search_tools` meta-tool and a pin that drops `structuredContent` (15). `tools/list` pages by 100 with `?paginated=true`, and logs and lists take cursors and filters. Controls on tool output size were not found (16). Errors carry an `error_type`, `data.reauth_reason`, a named guardrail on a block and a request id that matches the log row (17). The connectors endpoint returns annotation hints (read only, destructive) per tool. We could not confirm they appear in `tools/list`, and tool calls take no idempotency key (10). There is no Agent Handler SDK. Clients use the MCP SDKs or the Python CLI, and agent signup returns the ids a first call needs (10).\n- Security \u0026 auth 83: Access keys can be limited to named Registered Users and Tool Packs and to the `runtime:all` scope, with expiry, rotation and revocation, and a child key is never wider than its parent. The simplified URL uses OAuth. Two points off because the master key can't be revoked and agent signup returns an unrestricted key (28). Tool Packs select individual tools, input overrides fix field values and a runtime-only key can't change configuration. No confirmation step for destructive tools was found (15). The Security Gateway redacts or blocks tool arguments, and AI Guardrails can screen responses with content passed as data inside boundary markers. Guardrails fail open and new ones screen requests only (11). Tool Call Logs and an Audit Trail are kept 90 days by default with CSV export and an API. The SIEM log endpoints are Enterprise only (14). SOC 2 Type II, ISO 27001, HIPAA, yearly penetration tests and a disclosure programme at security@merge.dev. security.txt returned 404, no paid bounty terms were found, and the programme bars public disclosure without approval (15).\n- Payments \u0026 pricing 55: No x402, MPP or L402 (0). Plan prices are public, $1,000 a month for 25,000 credits up to $3,200 for 100,000, at one credit per tool call. The overage rate isn't published, so 15 of 20 (15). The Free plan gives 2,000 credits a month with no card (20). The docs and the OpenAPI file describe an unauthenticated `POST /api/v1/agent/signup/` that returns a key, a tool pack and a registered user. We did not call it, so this rests on the documentation. A paid plan needs a person and a card (20).\n- Task success: Pending. Task success needs the category task suites run through each tool, which haven't run yet, so this run doesn't score it. Its weight is shared across the assessed categories until then. A data provider's data-quality score is published on its listing now and becomes half of this category when it's scored.\n- Maintenance \u0026 community 69: Changelog entry for Agent Handler on 2 October 2026 and Merge CLI 0.5.0 on PyPI on 5 October (30). Agent Handler entries on 18 September, 25 September and 2 October, and CLI releases on 11 August, 9 September and 5 October (20). A closed service with a weekly changelog, a Discord community on Free, email support on Pro and tickets from the dashboard. The issue tracker linked from PyPI answered 404 (10). No Merge entry was found in the official MCP registry and there is no Agent Handler SDK. The CLI is current (5). The CLI supports Python 3.10 to 3.13 and is classified Beta. Its repository isn't public, so CI was not checked (4).\n- Transparency \u0026 trust 71: The editorial half. Closed service under a Master Services Agreement that names Merge API, Inc., a Delaware corporation, with California law (15). The MSA says Merge will not use Customer Data to train, fine-tune or improve any AI, and the privacy policy says it doesn't train generalised models on personal information. Logs are kept 90 days by default, a deleted Registered User's records go with it, and a closed account is purged after 30 days. Retention of stored credentials for the life of an account is left to the agreement and the DPA (22). No deprecation policy or dated sunset notices were found. The agreement can change on 30 days' notice or on continued use (4). The sub-processor list, updated 4 May 2026, gives each processor's location and includes Anthropic, OpenAI and AWS Bedrock. Standard plans are hosted on AWS US-East (17).\n\nFix list for a coding agent, everything this grade says the listing lacks, the biggest gain first (18 items): https://www.anchorterminal.com/fixes/merge-agent-handler.md (JSON https://www.anchorterminal.com/fixes/merge-agent-handler.json)\n\n### What we couldn't check\n\n- unchecked: live tool definitions from `tools/list`, including whether annotation hints are present, because reading them needs an account and we did not create one\n- unchecked: the unauthenticated agent signup endpoint, which we did not call. Its score rests on the docs and the OpenAPI file\n- unchecked: status history older than 30 days, which status.merge.dev did not show us\n- unchecked: trust.merge.dev and the Data Processing Addendum text, which we did not read in full\n- The numeric rate limits per Registered User and per organisation\n- The Pro overage rate per credit, and the Employees Pro price per user\n- Whether the Master Services Agreement or the Subscriber Agreement governs a self-serve Agent Handler account. Neither names the product\n- Why the repository linked from the CLI's PyPI page answers 404\n\n### Sources\n\n- product overview: \u003chttps://docs.merge.dev/merge-agent-handler/overview.md\u003e (seen 2026-10-08)\n- docs index for the product: \u003chttps://docs.merge.dev/merge-agent-handler/llms.txt\u003e (seen 2026-10-08)\n- docs index with text addressed to agents: \u003chttps://docs.merge.dev/llms.txt\u003e (seen 2026-10-08)\n- OpenAPI file: \u003chttps://docs.merge.dev/openapi/agent-handler-api.json\u003e (seen 2026-10-08)\n- MCP integration: \u003chttps://docs.merge.dev/merge-agent-handler/build/connecting-agents/mcp-integration.md\u003e (seen 2026-10-08)\n- agent quickstart and signup endpoint: \u003chttps://docs.merge.dev/merge-agent-handler/setup/agent-quickstart.md\u003e (seen 2026-10-08)\n- access keys: \u003chttps://docs.merge.dev/merge-agent-handler/administer/access-keys.md\u003e (seen 2026-10-08)\n- scoping access per user: \u003chttps://docs.merge.dev/merge-agent-handler/secure/scoping-access-per-user.md\u003e (seen 2026-10-08)\n- Security Gateway: \u003chttps://docs.merge.dev/merge-agent-handler/secure/security-gateway.md\u003e (seen 2026-10-08)\n- AI Guardrails: \u003chttps://docs.merge.dev/merge-agent-handler/secure/ai-guardrails.md\u003e (seen 2026-10-08)\n- billing and usage: \u003chttps://docs.merge.dev/merge-agent-handler/administer/billing-and-usage.md\u003e (seen 2026-10-08)\n- FAQ (GA status, hosting, retention, rate limits): \u003chttps://docs.merge.dev/merge-agent-handler/resources/faq.md\u003e (seen 2026-10-08)\n- troubleshooting and error types: \u003chttps://docs.merge.dev/merge-agent-handler/resources/troubleshooting.md\u003e (seen 2026-10-08)\n- audit trail and retention: \u003chttps://docs.merge.dev/merge-agent-handler/observe/audit-trail.md\u003e (seen 2026-10-08)\n- tool call logs: \u003chttps://docs.merge.dev/merge-agent-handler/observe/tool-call-logs.md\u003e (seen 2026-10-08)\n- Slack connector tools: \u003chttps://docs.merge.dev/merge-agent-handler/connectors/slack.md\u003e (seen 2026-10-08)\n- pricing: \u003chttps://www.merge.dev/pricing/agent-handler\u003e (seen 2026-10-08)\n- changelog: \u003chttps://www.merge.dev/changelog\u003e (seen 2026-10-08)\n- status page: \u003chttps://status.merge.dev\u003e (seen 2026-10-08)\n- Master Services Agreement: \u003chttps://www.merge.dev/legal/msa\u003e (seen 2026-10-08)\n- privacy policy: \u003chttps://www.merge.dev/legal/privacy-policy\u003e (seen 2026-10-08)\n- sub-processors: \u003chttps://www.merge.dev/legal/data-subprocessors\u003e (seen 2026-10-08)\n- security page: \u003chttps://www.merge.dev/security\u003e (seen 2026-10-08)\n- Merge CLI on PyPI: \u003chttps://pypi.org/project/merge-api/\u003e (seen 2026-10-08)\n- OAuth protected resource metadata: \u003chttps://ah-api.merge.dev/.well-known/oauth-protected-resource\u003e (seen 2026-10-08)\n- official MCP registry search: \u003chttps://registry.modelcontextprotocol.io/v0/servers?search=merge\u003e (seen 2026-10-08)\n\n## Who's behind it (provenance 83/100, checked 2026-10-08)\n\n| Check | Finding | Points |\n| --- | --- | --- |\n| Legal entity named | Merge API, Inc. | 20/20 |\n| Domain age | merge.dev, registered 2020-02-09 (6 years) | 11/15 |\n| Endpoint on the vendor's domain | ah-api.merge.dev | 15/15 |\n| Terms of service | read, states 6 of the 7 things a reader expects, and has 1 clause that costs points | 7.1/10 |\n| Privacy policy | read, states 8 of the 8 things a reader expects | 10/10 |\n| Status page | status.merge.dev | 10/10 |\n| Changelog | published | 10/10 |\n| security.txt | not found | 0/10 |\n\nThe Master Services Agreement, last updated 9 September 2026, names Merge API, Inc., a Delaware corporation at 353 Sacramento St., Floor 21, San Francisco, CA 94111, and applies to a customer who clicks to accept or uses the Service. merge.dev/legal/terms is a Terms of Use for the website.\n\nA separate Subscriber Agreement (last updated 2 April 2026) is written around End Customers and Merge Integrations, the Unified API's terms. Neither agreement names Agent Handler, so the MSA is our reading of which one governs it.\n\nThe privacy policy was last revised 20 August 2026. It leaves data processed for a customer to the customer agreement and the Data Processing Addendum.\n\nThe API and MCP server answer at ah-api.merge.dev and the dashboard at ah.merge.dev.\n\nwww.merge.dev/.well-known/security.txt returned 404. The security page gives security@merge.dev and a vulnerability disclosure programme.\n\nRDAP for merge.dev gives a registration date of 2020-02-09.\n\nstatus.merge.dev runs on Sorry and lists Merge Agent Handler as a component. It shows 30 days of notices.\n\n### Terms and privacy, as read\n\nA reading by a fixed set of rules, each answered with the vendor's own sentence. Not legal advice.\n\n**Terms of service** (https://www.merge.dev/legal/msa), read 2026-10-08, dated 2026-09-09, states 6 of the 7 things a reader expects.\n\n- To know. Restricts benchmarking or competitive use (costs points). \"Customer will not: (a) access (or allow a third party to access) the Service in order to benchmark, or monitor the availability, security, performance, or functionality of the Service, for any competitive purposes without Merge’s express written consent;\"\n- Gives the date it was last updated. Last updated 2026-09-09.\n- Names the governing law or courts. The law of California.\n- Says how changes to the terms are announced. Gives 30 days of notice before a change.\n- Not found in the text. Refers to a service level or uptime commitment.\n- Also in the text (2026-10-08). The customer indemnifies Merge against third-party claims related to any automated actions taken through the service at the customer's direction or configuration. \"related to: (a) Customer or a User engaging in a Prohibited Use; (b) Customer’s breach of Section 9.2 (Customer Representations \u0026 Warranties); and (c) any automated actions taken through the Service at Customer's direction or configuration.\"\n- Also in the text (2026-10-08). Merge says it will not use Customer Data, or allow any third party to use it, to train, fine-tune or improve any AI. \"Merge will not, and will not allow any third party to, use Customer Data to train, fine-tune, or improve any AI.\"\n- Also in the text (2026-10-08). Merge will delete Customer Data upon written request after the agreement terminates or expires. \"Upon termination or expiration of this Agreement for any reason, Merge will, upon written request, delete all Customer Data processed on behalf of Customer during the Subscription Term.\"\n\n**Privacy policy** (https://www.merge.dev/legal/privacy-policy), read 2026-10-08, dated 2026-08-20, states 8 of the 8 things a reader expects.\n\n- Gives the date it was last updated. Last updated 2026-08-20.\n- Says whether personal data is sold or shared for advertising. Says it does not sell personal data.\n- Gives a privacy contact. privacy@merge.dev.\n- Says where data is transferred or stored. Relies on standard contractual clauses.\n- Also in the text (2026-10-08). Merge states it does not develop, improve or train generalised AI or ML models using Personal Information. \"Merge does not develop, improve, or train generalized AI and/or ML models using Personal Information.\"\n- Also in the text (2026-10-08). Merge reserves the right to change the Privacy Policy without notice. \"We reserve the right to change this Privacy Policy from time to time without notice to you.\"\n\n## Live (updated 2026-10-09 10:14 UTC)\n\n- Right now: up, HTTP 401, 420 ms, checked 2026-10-09 10:14 UTC (mcp-initialize on `https://ah-api.merge.dev/mcp`, asks for auth)\n- Uptime 24h 100.0% (28 probes) · 30 days 100.0% (28 probes) · p50 449 ms · p95 1 s\n- Vendor status page: unknown, no machine-readable status found\n- Always current: https://www.anchorterminal.com/api/v1/live/merge-agent-handler.json\n\n## Probe metrics\n\nNot measured yet. Our benchmark probes haven't run, so there's no availability, latency or error rate from a run and Performance is pending. Live uptime, where we poll the endpoint, is under Live and doesn't change the score.\n\n## Prices\n\n| Item | Price | Unit | Note |\n| --- | --- | --- | --- |\n| Pro plan, 25,000 credits | $1000 | per month (plan) | One credit per tool call. 2,000 bonus credits. Overage billed at month end, rate not published |\n| Pro plan, 100,000 credits | $3200 | per month (plan) | Largest self-serve step on the pricing page |\n\nAcross all listings: https://www.anchorterminal.com/prices/index.md\n\n## Strengths\n\n- Access keys can be limited to named Registered Users and Tool Packs and to the `runtime:all` scope, with an expiry. A child key is never wider than its parent\n- The docs and OpenAPI file describe `POST /api/v1/agent/signup/`, which needs no credentials and returns a key, a tool pack and a registered user\n- The Free plan has 2,000 credits a month, one per tool call, and the quickstart says signup needs no card\n- Tool Call Logs record arguments, results, redactions and a latency breakdown per call, and every MCP result carries a request id that matches the log row\n- Changelog entries for Agent Handler are dated 18 September, 25 September and 2 October 2026\n\n## Weaknesses\n\n- No rate limit figures are published. The docs say only that MCP tool calls are limited per Registered User and per organisation, by the minute\n- Pro starts at $1,000 a month for 25,000 credits, and the overage rate isn't on the pricing page\n- Every `tools/call` spends a credit, including failed calls, calls blocked by a rule and calls made with a test key\n- AI Guardrails let a call through when the evaluator fails or is unreachable, and new guardrails screen requests only, by default\n- The master production key reaches every user and tool pack and can't be revoked, only regenerated. Agent signup returns an unrestricted key\n\n## Before you call it (notes for agents)\n\n1. Call `search_tools` with a plain-language intent before `tools/list`. A Tool Pack can hold hundreds of tools and Slack alone lists 50\n2. Add `?connectors=slack` to the MCP URL to limit a connection to named connectors, and `?paginated=true` only if the client follows `nextCursor`\n3. Carry a user-scoped key with `runtime:all` only. The production key reaches every Registered User and the management endpoints\n4. On `reauth_required` read `data.reauth_reason`, then call `authenticate_\u003cconnector\u003e` by name. Intent search doesn't return the authenticate tools\n5. Pin `MCP-Protocol-Version` to `2025-03-26` when context is tight. Later revisions repeat object results in `structuredContent`\n\n## Connect\n\nInstall:\n\n```bash\npipx install merge-api\n```\n\nFirst request:\n\n```bash\ncurl -X POST \"https://ah-api.merge.dev/api/v1/tool-packs/$TOOL_PACK_ID/registered-users/$REGISTERED_USER_ID/mcp\" \\\n -H \"Authorization: Bearer $MERGE_AGENT_HANDLER_API_KEY\" -H \"Content-Type: application/json\" \\\n -H \"Accept: application/json, text/event-stream\" \\\n -d '{\"jsonrpc\":\"2.0\",\"id\":1,\"method\":\"tools/list\"}'\n```\n\nClaude Code:\n\n```bash\nclaude mcp add --transport http agent-handler https://ah-api.merge.dev/mcp\n```\n\nMCP client configuration:\n\n```json\n{\n \"mcpServers\": {\n \"agent-handler\": {\n \"headers\": {\n \"Authorization\": \"Bearer \\u003cYOUR_API_KEY\\u003e\"\n },\n \"url\": \"https://ah-api.merge.dev/api/v1/tool-packs/\\u003cTOOL_PACK_ID\\u003e/registered-users/\\u003cREGISTERED_USER_ID\\u003e/mcp\"\n }\n }\n}\n```\n\nThrough letme (picks today, calling later): https://letme.dev/merge-agent-handler. letme answers with the pick and how to call it direct; calling through letme (one key, the vendor's own price) comes later. How it works: https://www.anchorterminal.com/letme/index.md\n\n## Similar tools\n\nRanked by shared capabilities, then score. Same-category tools with no shared capability key are listed last.\n\n| Tool | Grade | Score | Rank | Shared capabilities | x402 | Markdown |\n| --- | --- | --- | --- | --- | --- | --- |\n| Composio (API + MCP) | BB | 75.1 | 54 | automation.apps, automation.auth, automation.actions, agent.tools | no | https://www.anchorterminal.com/tools/composio-rube.md |\n| Zapier MCP (agent actions) | C | 58.1 | 531 | automation.apps, automation.auth, automation.actions, agent.tools | no | https://www.anchorterminal.com/tools/zapier-mcp.md |\n| Pipedream API + MCP | B | 65.5 | 288 | automation.apps, automation.auth, agent.tools | no | https://www.anchorterminal.com/tools/pipedream.md |\n| Workato API + MCP | C | 58 | 533 | automation.apps, automation.auth, agent.tools | no | https://www.anchorterminal.com/tools/workato.md |\n| Tray.ai API + MCP | C | 55.5 | 589 | automation.apps, automation.auth, agent.tools | no | https://www.anchorterminal.com/tools/tray.md |\n| Paragon ActionKit + MCP | D | 47.5 | 741 | automation.apps, automation.auth, agent.tools | no | https://www.anchorterminal.com/tools/paragon.md |\n\n## Panel reviews (0)\n\nReviewed by the Anchor panel (https://www.anchorterminal.com/reviewers/index.md): .\n\nDesk reviews, written from public documentation, pricing, terms, source and status history on 1 October 2026. No calls made. For a desk review, the outcome says whether the reviewer's questions could be answered from public material: success, partial or failure. How reviews work: https://www.anchorterminal.com/reviews/how-it-works.md\n\n## Notable\n\n- The overview counts 190+ connectors as of September 2026, split between connectors Merge builds and generic connectors that are MCP servers run by outside vendors (source: \u003chttps://docs.merge.dev/merge-agent-handler/connectors/overview.md\u003e)\n- Access keys take Tool Pack and Registered User restrictions, two scopes (`runtime:all`, `management:all`) and an expiry, and a key can only mint, rotate or revoke keys its own grant covers (source: \u003chttps://docs.merge.dev/merge-agent-handler/administer/access-keys.md\u003e)\n- The Security Gateway scans tool arguments with Presidio-based entity rules and custom regex (allow, redact or block), and AI Guardrails judge requests and, when switched on, responses (source: \u003chttps://docs.merge.dev/merge-agent-handler/secure/security-gateway.md\u003e)\n- AI Guardrails fail open. When the evaluator is unreachable or returns no verdict the call proceeds and an evaluation-failure row is written (source: \u003chttps://docs.merge.dev/merge-agent-handler/secure/ai-guardrails.md\u003e)\n- The agent quickstart and the OpenAPI file describe an unauthenticated signup endpoint, rate limited per IP, that returns an unrestricted production key once and a claim link for a person (source: \u003chttps://docs.merge.dev/merge-agent-handler/setup/agent-quickstart.md\u003e)\n- The docs carry text addressed to AI agents, including a 'Rules for agents' list in the agent quickstart and a line in llms.txt telling an agent how to create an account. We read it as data and did not act on it (source: \u003chttps://docs.merge.dev/llms.txt\u003e)\n- llms.txt gives the OpenAPI file as `openapi/agent-handler.json`, which answered 404 on 8 October 2026. The file is at `openapi/agent-handler-api.json` (source: \u003chttps://docs.merge.dev/openapi.json\u003e)\n- The Master Services Agreement bars access to benchmark or monitor availability, security, performance or functionality for competitive purposes without Merge's written consent (source: \u003chttps://www.merge.dev/legal/msa\u003e)\n\n## Compare\n\n- [letme vs Merge Agent Handler](https://www.anchorterminal.com/compare/letme-vs-merge-agent-handler.md): F 36.4 vs B 69.5\n- [Composio (API + MCP) vs Merge Agent Handler](https://www.anchorterminal.com/compare/composio-rube-vs-merge-agent-handler.md): BB 75.1 vs B 69.5\n- [Merge Agent Handler vs Zapier MCP (agent actions)](https://www.anchorterminal.com/compare/merge-agent-handler-vs-zapier-mcp.md): B 69.5 vs C 58.1\n\n## Verify this listing\n\nFor the vendor. The badge or a plain link to this page verifies the listing, from a page on merge.dev or one of its subdomains. It shows the listing is the vendor's and that the vendor knows it's here, and it never changes a grade, rank or review. The vendor sends the page's address to `POST https://www.anchorterminal.com/api/v1/verify` as `{\"slug\": \"merge-agent-handler\", \"url\": \"…\"}`, or calls the `verify_listing` tool at https://www.anchorterminal.com/mcp. We fetch the page once, then again every week; two failed checks in a row and the verification lapses, and a later pass restores it. What we check: https://www.anchorterminal.com/builders/index.md#verify\n\nHTML badge:\n\n```html\n\u003ca href=\"https://www.anchorterminal.com/tools/merge-agent-handler\"\u003e\u003cimg src=\"https://www.anchorterminal.com/badges/merge-agent-handler.svg\" alt=\"Merge Agent Handler on Anchor Terminal\" height=\"20\"\u003e\u003c/a\u003e\n```\n\nMarkdown badge, for a README:\n\n```markdown\n[![Merge Agent Handler on Anchor Terminal](https://www.anchorterminal.com/badges/merge-agent-handler.svg)](https://www.anchorterminal.com/tools/merge-agent-handler)\n```\n\nPlain link:\n\n```html\n\u003ca href=\"https://www.anchorterminal.com/tools/merge-agent-handler\"\u003eMerge Agent Handler on Anchor Terminal\u003c/a\u003e\n```\n\n## Share this listing\n\nFor the vendor. Sharing assets for social media, two PNGs of 1200 × 630 that say Merge Agent Handler is listed on Anchor Terminal, with the vendor's logo and this page's address and no grade or score.\n\n- Dark: https://www.anchorterminal.com/assets/share/merge-agent-handler-dark.png\n- Light: https://www.anchorterminal.com/assets/share/merge-agent-handler-light.png\n", "meta": { "attribution": "Anchor Terminal (https://www.anchorterminal.com)", "docs": "https://www.anchorterminal.com/docs/", "generatedAt": "2026-10-09", "license": "CC-BY-4.0", "method": "https://www.anchorterminal.com/benchmark/", "methodology": "0.4", "openapi": "https://www.anchorterminal.com/openapi.json", "preview": false, "run": "2026-10-01", "runLabel": "October 2026 research run" }, "page": { "breadcrumbs": [ { "name": "Home", "url": "https://www.anchorterminal.com/" }, { "name": "Terminal", "url": "https://www.anchorterminal.com/tools/" }, { "name": "Agent tool access", "url": "https://www.anchorterminal.com/categories/aggregator" }, { "name": "Merge Agent Handler", "url": "" } ], "description": "Merge Agent Handler is a hosted MCP server from Merge API, Inc. that gives AI agents tools across more than 190 third-party connectors, with per-user credentials, tool packs, a security gateway and logs of every call.", "facts": [ "rank #174 of 842", "OAuth or key auth", "0 desk reviews" ], "h1": "Merge Agent Handler", "image": "https://www.anchorterminal.com/assets/og/tools-merge-agent-handler.png", "path": "/tools/merge-agent-handler", "published": "2026-10-01", "section": "tools", "title": "Merge Agent Handler review for AI agents, grade B (69.5/100)", "toc": null, "updated": "2026-10-09", "url": "https://www.anchorterminal.com/tools/merge-agent-handler" }, "tokens": { "markdown": 8000, "slim": 1930 }, "version": 1 }