agentboards.org
Compare/AutoGen vs GenericAgent

AutoGenvsGenericAgent

Generated from the two spec rows. Green marks the better value where a spec has a clear direction. Everything else is just different.

AutoGen
Microsoft · Agent framework
OSSMCP
Panel
4.5
7 spec wins
Reliability
4.8
Usefulness
4.3
Cost
6.3
Longevity
2.7

“The docs are CC-BY-4.0, so you may freely redistribute the notice telling you to use something else.”

GenericAgent
lsdefine · Agent framework
OSS
Panel
4.4
2 spec wins
Reliability
3.2
Usefulness
4.7
Cost
5.5
Longevity
4.3

“Grants an LLM full system control over a local computer without a sandbox.”

Spec by spec

SpecAutoGenGenericAgent
Architecture
CategoryAgent frameworkAgent framework
Runslocallocal
Platformsmacos, linux, windowsmacos, linux, windows
Context windownot documentednot documented
Protocols
MCP clientYesNo
MCP serverNoNo
Capabilities
Runs terminal commandsYesYes
Multi-file editsNoYes
Git operationsNoYes
Browser controlYesYes
Sandboxed executionYesNo
Multi-agent orchestrationYesNo
Headless / CI modeYesNo
Models
BackboneOpenAI, Azure OpenAI, any autogen-ext model clientClaude, Gemini, Kimi, MiniMax
Bring your own modelYesYes
Local modelsYesNo
Cost
Pricing modelbyokbyok
Starts at$0/mon/a
Free tierYesNo
Bring your own keyYesYes
Openness
Open sourceYesYes
LicenseMITunknown
GitHub stars61,25314,276

Which one would each critic pick

CriticAutoGenGenericAgentPick
El Juez——not enough reviews
El Amigo4.55.5GenericAgent — GenericAgent is a powerful framework for agent research, but its direct control over your system without a sandbox makes it too risky for daily development work.
El Crítico4.05.8GenericAgent — This agent gets full system control without a sandbox, making every mistake a potential security incident.
El Profesor5.05.0no preference
La Inversora3.83.8no preference
La Jefa4.02.0AutoGen — As a dependency it has no vendor support, no roadmap and no SLA, which for sixty engineers means a migration project we did not budget; not yet, and plan the exit.
El Hacker6.04.5AutoGen — MIT code, McpWorkbench for MCP, any autogen-ext model client including local ones, and the fork already happened: the original maintainers ship the 0.2 line as AG2 under Apache-2.0.

Picks are derived from each critic's own scores. Humans vote on matchups on the duels page.

Want a third column? The compare tool handles any two agents. Three-way comparisons are on the roadmap once the spec rows are all verified.