The unified server can discover and execute many ACI.dev managed functions, so restrict permissions before connecting it to an autonomous agent., App-specific servers should be preferred when a workflow only needs a small set of integrations., Connected apps may read, create, update, delete, or send data depending on the selected functions and linked account permissions., Require human approval before email, calendar, CRM, ticketing, infrastructure, billing, deployment, or account-administration actions.
Privacy notes
Tool names, function arguments, prompts, linked account identifiers, execution results, logs, and connected-service data may pass through ACI.dev, the selected third-party apps, the MCP client, and the model provider., API keys and linked-account credentials must be protected in client config, environment files, logs, screenshots, and shared prompts., Review ACI.dev account, tenant, retention, audit-log, and OAuth settings before routing production or customer data through the server.
Author
Aipolabs
Submitted by
oktofeesh1
Claim status
unclaimed
Last verified
2026-06-05
Decision playbook
Review trust signals before you adopt
Signals are present but mixed. Use the checklist below to confirm the source and operational safety for your environment.
Compare context
Selected
0
Current score
63
Baseline
—
Delta
No baseline selected
No major trust-signal divergence detected in the current selection.
Source and provenance checks
Needs review
Confirm ownership and provenance before trusting install instructions.
Source link availableRequired
Open the canonical repository and verify ownership.
Done
Source provenance statusRequired
Marked as source-backed.
Done
Metadata reviewed
No reviewed flag detected in metadata.
Pending
Safety and privacy checks
Complete
Validate risk disclosures before installation or API wiring.
Safety notes presentRequired
Review the listed safety guidance before running commands.
Done
Privacy notes presentRequired
Review data handling notes before connecting accounts or secrets.
Done
Trust level risk gateRequired
Trust level does not block evaluation.
Done
Package and install checks
Needs review
Check package metadata and artifact integrity signals.
Install payload available
Install or copy payload is available for review.
Done
Package verification flag
No package verification flag provided.
Pending
Checksum metadata
No checksum provided for downloaded artifact.
Pending
Compare-driven decision checks
Needs review
Use compare context to validate trade-offs before adoption.
Compare tray has multiple entries
Add at least one more entry to compare trust differences.
4 safety and 3 privacy notes across 4 risk areas. Review closely: credentials & tokens, permissions & scopes, third-party handling.
4 areas
SafetyPermissions & scopesThe unified server can discover and execute many ACI.dev managed functions, so restrict permissions before connecting it to an autonomous agent.
SafetyGeneralApp-specific servers should be preferred when a workflow only needs a small set of integrations.
SafetyPermissions & scopesConnected apps may read, create, update, delete, or send data depending on the selected functions and linked account permissions.
SafetyPermissions & scopesRequire human approval before email, calendar, CRM, ticketing, infrastructure, billing, deployment, or account-administration actions.
PrivacyThird-party handlingTool names, function arguments, prompts, linked account identifiers, execution results, logs, and connected-service data may pass through ACI.dev, the selected third-party apps, the MCP client, and the model provider.
PrivacyCredentials & tokensAPI keys and linked-account credentials must be protected in client config, environment files, logs, screenshots, and shared prompts.
PrivacyCredentials & tokensReview ACI.dev account, tenant, retention, audit-log, and OAuth settings before routing production or customer data through the server.
Disclosure: editorial
Safety notes
The unified server can discover and execute many ACI.dev managed functions, so restrict permissions before connecting it to an autonomous agent.
App-specific servers should be preferred when a workflow only needs a small set of integrations.
Connected apps may read, create, update, delete, or send data depending on the selected functions and linked account permissions.
Require human approval before email, calendar, CRM, ticketing, infrastructure, billing, deployment, or account-administration actions.
Privacy notes
Tool names, function arguments, prompts, linked account identifiers, execution results, logs, and connected-service data may pass through ACI.dev, the selected third-party apps, the MCP client, and the model provider.
API keys and linked-account credentials must be protected in client config, environment files, logs, screenshots, and shared prompts.
Review ACI.dev account, tenant, retention, audit-log, and OAuth settings before routing production or customer data through the server.
Prerequisites
ACI.dev account, API key, and linked account owner ID.
Connected app credentials or linked accounts for the tools the agent should use.
MCP client that can launch Python package commands with environment variables.
Clear allowlist of apps, functions, and user accounts before exposing write-capable tools.
ACI MCP Servers provide MCP access to ACI.dev managed functions. The package
offers a unified server with search-and-execute meta tools and an app-specific
server for exposing functions from selected apps.
This entry points at the open-source aipotheosis-labs/aci-mcp package rather
than the broader ACI.dev platform repository. The upstream README describes two
server modes: aci-mcp-unified for searching and executing available functions,
and aci-mcp-apps for direct access to functions from explicitly selected apps.
These sources were reviewed on 2026-06-05. Prefer the live ACI.dev docs for
current API key, linked account owner, app selection, client configuration, and
deployment behavior.
Features
Unified MCP server with search and execution tools for ACI.dev functions.
App-specific MCP server for exposing selected app functions directly.
PyPI package runnable with uvx.
Docker usage documented by the upstream project.
ACI.dev-managed authentication, linked accounts, app integrations, and
function execution.
Documentation for unified and app-specific MCP client integration.
For narrower access, use the upstream apps-server mode and pass only the apps
needed for the workflow.
Use Cases
Give Claude one MCP connection for many ACI.dev-managed integrations.
Use the unified server to discover relevant functions during exploratory
workflows.
Use the app-specific server for safer, narrower access to approved apps.
Prototype assistants that need calendars, email, search, CRM, ticketing, or
developer-platform tools behind a common MCP interface.
Compare unified tool discovery with explicit function allowlisting before
production use.
Safety and Privacy
ACI MCP Servers can expose high-impact third-party actions through one MCP
connection. Prefer the smallest useful app set, review linked account scopes,
and require human approval for write, send, deployment, billing, admin, or
destructive operations.
Treat ACI API keys, linked account IDs, prompts, function arguments, execution
results, OAuth data, logs, and connected-service records as sensitive. Review
ACI.dev account controls and connected app permissions before using the server
with production or customer data.
Duplicate Check
No aipotheosis-labs/aci-mcp entry or source URL was found in content/mcp.
This is distinct from individual app MCP servers because it focuses on ACI.dev's
unified and app-specific function gateway rather than one service-specific API.
Disclosure
Editorial listing. No paid placement or affiliate link is used. The MCP package
is open source, but the workflow depends on an ACI.dev account and managed
function infrastructure.
Show that ACI MCP Servers is listed on HeyClaude. Paste this Markdown into your README — it renders the badge and links back to this page.
[](https://heyclau.de/entry/mcp/aci-mcp-servers)
How it compares
ACI MCP Servers side by side with 3 alternatives on trust, install, platform support, and disclosed safety notes — all from reviewed registry metadata.
1 trust signal differ across this comparison (Submitter).
Open-source MCP router from Klavis AI that progressively exposes actions from multiple configured MCP servers instead of flooding Claude with every tool.
Official MCP server for agent-device, Callstack's device automation CLI for inspecting, controlling, debugging, recording, and collecting evidence from iOS, Android, TV, macOS, Linux, React Native, Expo, Flutter, and native apps.
Remote Agentage Memory MCP server giving Claude one shared, portable memory across every AI you use - search, read, write, edit, list, and delete plain Markdown notes you own, over Streamable HTTP with OAuth 2.1.
✓The unified server can discover and execute many ACI.dev managed functions, so restrict permissions before connecting it to an autonomous agent.
App-specific servers should be preferred when a workflow only needs a small set of integrations.
Connected apps may read, create, update, delete, or send data depending on the selected functions and linked account permissions.
Require human approval before email, calendar, CRM, ticketing, infrastructure, billing, deployment, or account-administration actions.
✓Strata can execute tools on every enabled downstream MCP server; treat it as a privileged router, not a harmless discovery layer.
Review each downstream server before adding it, because Strata can route file, browser, database, SaaS, shell, or write-capable actions through one interface.
The HTTP/SSE mode starts an ASGI server and the source currently binds uvicorn to all interfaces; restrict network exposure with firewall rules, reverse proxies, or local-only deployment.
Store bearer tokens, OAuth credentials, headers, and environment variables in a secret manager or private config file rather than committing Strata configuration.
Disable or remove servers that should not be available to a specific project, user, or model session.
✓Agent Device MCP exposes structured tools backed by `AgentDeviceClient`; the docs state it does not expose generic shell execution over MCP.
Tools and CLI workflows can open apps, inspect UI, tap, type, scroll, perform gestures, wait, assert state, handle alerts, and close sessions.
Evidence workflows can capture screenshots, recordings, logs, traces, network traffic, performance samples, crash context, React profiles, and replay files.
Mutating commands should run serially against one session, and separate sessions or devices should be used for parallel work.
Prefer dedicated test devices or simulators, and require approval before entering credentials, submitting forms, changing settings, installing apps, sending messages, or touching production accounts.
✓The memory__write, memory__edit, and memory__delete tools create, modify, and remove your stored notes; review actions before granting broad access.
Privacy notes
✓Tool names, function arguments, prompts, linked account identifiers, execution results, logs, and connected-service data may pass through ACI.dev, the selected third-party apps, the MCP client, and the model provider.
API keys and linked-account credentials must be protected in client config, environment files, logs, screenshots, and shared prompts.
Review ACI.dev account, tenant, retention, audit-log, and OAuth settings before routing production or customer data through the server.
✓Strata stores configured server names, commands, URLs, headers, and environment variables in a local JSON configuration file.
User prompts, action discovery queries, tool schemas, tool arguments, and downstream server responses can all become visible to the MCP client and model provider.
Routed tools may expose third-party account data, files, repositories, browser state, database rows, or workspace context depending on which servers are enabled.
Hosted Klavis APIs and SDKs are separate from the local open-source router; review Klavis account, API-key, and data-handling terms before using hosted instances.
✓Screenshots, recordings, traces, logs, network dumps, replay files, reports, UI snapshots, typed input, and React profiles can contain private UI state, tokens, request data, customer information, or credentials.
macOS, iOS, Android, and TV automation can expose local app state, notifications, device names, package identifiers, app content, system dialogs, and permission prompts.
Network inspection artifacts may include headers, payloads, session identifiers, URLs, and API data; review before sharing or committing.
Interactive CLI runs may check npm for newer package versions unless `AGENT_DEVICE_NO_UPDATE_NOTIFIER=1` is set.
✓Notes you read or write are stored server-side in your Agentage Memory and mirrored locally as Markdown you can export anytime; note contents pass through model context.
Prerequisites
ACI.dev account, API key, and linked account owner ID.
Connected app credentials or linked accounts for the tools the agent should use.
MCP client that can launch Python package commands with environment variables.
Clear allowlist of apps, functions, and user accounts before exposing write-capable tools.
Python 3.10 or newer, matching the open-strata package metadata.
pipx or pip available on the machine that will run the router.
One or more MCP servers you already trust and are authorized to connect.
Any API keys, OAuth credentials, headers, or environment variables required by the downstream MCP servers.
Node.js 22 or newer and `agent-device` installed globally or project-locally.
Xcode tooling for iOS, tvOS, or macOS targets, or Android SDK and ADB for Android targets.
Device, simulator, emulator, TV, macOS, or Linux desktop target that the agent is allowed to automate.
Required local permissions such as Android device trust, iOS Developer Mode, macOS Accessibility, and Screen Recording where applicable.
agentage account (sign in at memory.agentage.io)
OAuth 2.1 authentication (PKCE + Dynamic Client Registration, handled in the connect flow)
Network access to memory.agentage.io (HTTPS required)
Claude Desktop 0.7.0+ or Claude Code with MCP (HTTP transport) support