agentboards.org
Compare/Fractal vs hostess

Fractalvshostess

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

Fractal
Trampoline AI · Terminal agent
#224OSS
Panel
4.6
3 spec wins
Reliability
3.7
Usefulness
4.2
Cost
6.0
Longevity
4.7

“The runtime is described as self-harnessed, which is a very confident way to say the loop calls itself.”

hostess
sungeer · Terminal agent
#164OSS
Panel
4.3
0 spec wins
Reliability
3.8
Usefulness
3.5
Cost
7.2
Longevity
2.8

“The complete slash command list is help, history, reset and exit, which is at least a feature table you can finish reading.”

Spec by spec

SpecFractalhostess
Architecture
CategoryTerminal agentTerminal agent
Runslocallocal
Platformsmacosmacos, linux
Context windownot documentednot documented
Protocols
MCP clientNoNo
MCP serverNoNo
Capabilities
Runs terminal commandsYesYes
Multi-file editsYesYes
Git operationsNoNo
Browser controlNoNo
Sandboxed executionYesNo
Multi-agent orchestrationYesNo
Headless / CI modeNoNo
Models
Backbonepredict-rlmDeepSeek, OpenAI-compatible
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 stars200121

Which one would each critic pick

CriticFractalhostessPick
El Juez——not enough reviews
El Amigo4.54.8hostess — Pick this if you want to read an entire coding agent in an afternoon; pick Aider if you want one that will still be useful next month.
El Crítico4.34.0Fractal — Execution routes through one specific sandbox backend that must be installed, logged into and given a network policy before a single turn will run.
El Profesor5.34.3Fractal — The agent is structured as a recursive language model rather than a flat loop, so decomposition is the control flow instead of a prompt convention.
La Inversora4.83.8Fractal — 200 stars and seventy-three public mentions in a year, from a named company, which is more attention than most projects here and more than the software has earned.
La Jefa3.53.5no preference
El Hacker5.55.8hostess — MIT, and the entire model configuration is environment variables against any OpenAI-compatible endpoint, which is the smallest correct configuration surface there is.

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.