agentboards.org
Compare/Cowork Forge vs Looper

Cowork ForgevsLooper

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

Cowork Forge
sopaco · Autonomous SWE
#258OSS
Panel
4.7
0 spec wins
Reliability
4.0
Usefulness
4.3
Cost
6.5
Longevity
4.0

“The engineer agent writes the code and then writes the delivery report, which is the most realistic simulation of a development team yet.”

Looper
nexu-io · Autonomous SWE
#216OSS
Panel
5.6
2 spec wins
Reliability
5.5
Usefulness
5.8
Cost
6.2
Longevity
4.8

“It only picks up issues that are both assigned and labelled, a bar most backlogs have never once cleared.”

Spec by spec

SpecCowork ForgeLooper
Architecture
CategoryAutonomous SWEAutonomous SWE
Runslocallocal
Platformsmacos, linuxmacos, linux
Context windownot documentednot documented
Protocols
MCP clientNoNo
MCP serverNoNo
Capabilities
Runs terminal commandsYesYes
Multi-file editsYesYes
Git operationsYesYes
Browser controlNoNo
Sandboxed executionNoNo
Multi-agent orchestrationYesYes
Headless / CI modeNoYes
Models
BackboneClaude Code, Codex, OpenCodemultiple providers
Bring your own modelYesYes
Local modelsNoNo
Cost
Pricing modelbyokbyok
Starts at$0/mo$0/mo
Free tierYesYes
Bring your own keyYesYes
Openness
Open sourceYesYes
LicenseMITMIT
GitHub stars97128

Which one would each critic pick

CriticCowork ForgeLooperPick
El Juez——not enough reviews
El Amigo5.05.5Looper — Pick this if your backlog is full of small, well-specified issues; pick a supervised terminal agent if your tickets need a conversation before they need code.
El Crítico4.04.3Looper — Each agent loops against its own success criteria rather than a step count, and the fixer and reviewer ping-pong until a pull request converges, with no stated ceiling.
El Profesor5.06.3Looper — The forge is designated the source of truth and the planner emits a spec pull request before code, so the intermediate artefact is a reviewable document rather than hidden state.
La Inversora4.55.0Looper — A hundred and thirteen stars and no company, entering the one segment where the competitors are venture-funded and sell the same promise with a support contract.
La Jefa4.05.3Looper — No seat cost across sixty engineers, and it runs as a daemon on infrastructure we own, which means it needs a service account with commit rights to our repositories.
El Hacker5.87.3Looper — MIT, `go install` and a single binary, and it speaks Forgejo as well as GitHub, so the whole loop runs on a forge I host myself.

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.