RepoRadar
Leave a Tip
Item detail

bytedance/deer-flow

bytedance/deer-flow is an MIT-licensed, long-horizon SuperAgent harness from ByteDance that researches, codes, and creates with sandboxed execution, persistent memory, and tool integration so multi-agent AI workflows can run multi-day tasks without constant human babysitting.

Open source →Back to radar
Score8.6
Popularity8.7
Risklow
TierGold
Score breakdown
Usefulness8.0
Novelty8.0
Momentum9.0
Maturity6.9
Open-source/build8.4
Evidence7.2
Workflow potential10.0
Setup ease6.4

Popularity is tracked separately. Support, ads, sponsorships, and tips never affect these signals.

Why it matters

Useful for AI builders, research teams, and developers who want a self-hostable, long-horizon agent harness that can chain research, coding, and creative work in a single sandboxed run, instead of stitching together a hand-rolled multi-agent loop on top of an LLM API.

Who should use it

AI builders who want a long-horizon agent harness instead of a single-turn task runnerresearch teams who want a self-hostable SuperAgent that can chain research, coding, and creative workdevelopers who need sandboxed execution and persistent memory for multi-day agent workflowsopen-source contributors who want a well-maintained MIT-licensed alternative to closed-source agent SaaS

Who should skip it

Skip if the source link, docs, or setup requirements do not match your workflow.

Risk explanation

It runs sandboxed code on your behalf and is intended to back long-horizon agent workflows, so scope the sandbox permissions, audit what gets persisted in agent memory, and confirm sensitive intermediate files are wiped between runs before promotion.

Evidence links

Closest alternatives / related signals

agentsuperagentlong-horizonorchestrationsandboxmemoryopen-sourcebytedance