AgentSidekick

agent-device vs Argent

agent-device is a MCP server from callstackincubator (MIT licence, local hosting, 5.5k GitHub stars, last push 2026-11-21); Argent is a MCP server from software-mansion (Apache-2.0 licence, local hosting, 3.2k GitHub stars, last push 2026-11-21). They share 4 of 7 capabilities and integrations identified for either (overlap 0.57); only agent-device covers E2E Testing, Android Development; only Argent covers Performance Review. Figures as of 2026-11-21. DER

agent-device profileArgent profileOpen in the compare tool

Side-by-side facts

7 of 16 rows differ (marked). Unknown means no source we track states it.

Fieldagent-deviceArgent
TypeMCP server · mcp-local SRCMCP server · mcp-local SRC
Publisher · differscallstackincubator SRCsoftware-mansion SRC
Licence · differsMIT SRCApache-2.0 SRC
Open sourceYes DERYes DER
Pricing modelUnknownUnknown
HostingLocal (runs on your machine) SRCLocal (runs on your machine) SRC
Latest version · differs0.16.9 SRC0.34.0 SRC
Languagestypescript SRCtypescript SRC
GitHub stars · differs5.5k (5,473) SRC3.2k (3,223) SRC
Forks · differs422 SRC160 SRC
Last push2026-11-21 SRC2026-11-21 SRC
ArchivedNo SRCNo SRC
First published · differs2026-05-11 SRC2026-08-10 SRC
Versions recorded · differs6 SRC2 SRC
VerificationListed by the publisher in a source registry SRCListed by the publisher in a source registry SRC
Last verifiedUnknownUnknown

Shared and unique capabilitiesDER

Capabilities are assigned by keyword rules from each description unless a researcher set them; integrations are services named in the description or package metadata.

Platform compatibility

Status per platform from the evidence we hold. No pairing has been tested by AgentSidekick.

Platformagent-deviceArgent
ContinueProtocol-levelProtocol-level
OpenCodeProtocol-levelProtocol-level
Claude CodeProtocol-levelProtocol-level
CodexProtocol-levelProtocol-level
Gemini CLIProtocol-levelProtocol-level
ZedProtocol-levelProtocol-level
ClineProtocol-levelProtocol-level
WarpProtocol-levelProtocol-level
gooseProtocol-levelProtocol-level
Qwen CodeProtocol-levelProtocol-level
KiroProtocol-levelProtocol-level
Amazon Bedrock AgentCoreProtocol-levelProtocol-level
AmpProtocol-levelProtocol-level
Apps in ChatGPT / Apps SDK (now ChatGPT Plugins)Protocol-levelProtocol-level
Augment CodeProtocol-levelProtocol-level
ChatGPTProtocol-levelProtocol-level
Claude (claude.ai apps)Protocol-levelProtocol-level
Claude DesktopProtocol-levelProtocol-level
CursorProtocol-levelProtocol-level
DevinProtocol-levelProtocol-level
Devin Desktop (formerly Windsurf)Protocol-levelProtocol-level
Factory (Droid)Protocol-levelProtocol-level
GitHub CopilotProtocol-levelProtocol-level
GitHub Copilot cloud agent (formerly Copilot coding agent)Protocol-levelProtocol-level
Google AntigravityProtocol-levelProtocol-level
JetBrains AI AssistantProtocol-levelProtocol-level
JunieProtocol-levelProtocol-level
LovableProtocol-levelProtocol-level
TabnineProtocol-levelProtocol-level

18 other platform(s): Unknown or Platform discontinued for both.

Maintenance

Repository activity read from the GitHub API. Activity is not quality.

Measureagent-deviceArgent
StatusActive DERActive DER
Days since last push0 DER0 DER
Open issues154 SRC17 SRC
Latest GitHub releasev0.21.41 · 2026-11-19 SRCv0.34.0 · 2026-11-18 SRC
Newest registry version0.16.9 · 2026-06-01 SRC0.34.0 · 2026-11-19 SRC

Evaluation dimensions

Separate signals with their evidence; there is no overall score. Hover or focus a status for its evidence.

Dimensionagent-deviceArgent
Source authorityPublisher-submittedPublisher-submitted
Publisher verificationNamespace verified by registryNamespace verified by registry
Maintenance activityActiveActive
Licence clarity · differsMITApache-2.0
Installation clarityMachine-readableMachine-readable
DocumentationRepository / website onlyRepository / website only
Compatibility evidenceProtocol-level onlyProtocol-level only
Security reviewNot reviewedNot reviewed

Which fits which requirementDER

Derived from the differences above, one rule per line. Nothing here says either product is better; check the requirement against your own needs.

Choose agent-device if…

  • you need E2E Testing, Android Development (classified for agent-device only)

Choose Argent if…

  • you need Performance Review (classified for Argent only)

Sources

Report incorrect information

Page data last updated .