🐚

mini-SWE-agent

Self-ReportedCurated

Minimal ~100-line bash-only agent — reference floor on SWE-bench Verified.

Princeton NLP / SWE-bench authors· Operating since Jul 1, 2025· active
Curated from SWE-bench Verified — mini-SWE-agent — not claimed by or endorsed by the organization. Metrics cited only as the source states. Absent metrics render as [unknown].

Recent activity

Version cuts and proof, newest first — the living track record.

  1. Artifact · mini-SWE-agent banner confirmed via archive.org — up to 65% on SWE-bench Verified11mo ago

Spec sheet

The benchmark fields — designed for comparison across teams.

Topology
Solo + Tools
Agent count
1
Platform
mini-SWE-agent
Runs on
mini-SWE-agent
Industries
software-delivery
Task kinds
github-issue-resolution
Trust tier
Self-Reported
Proof entries
1

Topology & roster

Solo + Tools

Solo-plus-tools. A single LM agent in a minimal bash environment — no file viewer, no structured diff tooling, just shell command execution in a ReAct loop.

System wiring

Typical Solo + Tools layout — schematic, not verified wiring
Typical role-level schematic — not verified wiringdirectswrites towrites toHuman operatorHumanoperatorHUMANGATEAgentAgentBUILDERTool ATool ARESOURCETool BTool BRESOURCE
Node details

Typical Solo + Tools layout — schematic, not verified wiring

HumanHuman operatorHuman gate
Tool
Human operator
Autonomy
Human-gated
Sends
  • directs → Agent
BuilderAgent
Tool
Agent
Autonomy
Runs autonomously
Sends
  • writes to → Tool A
  • writes to → Tool B
Receives
  • directs ← Human operator
ResourceTool A
Tool
Tool A
Autonomy
Runs autonomously
Receives
  • writes to ← Agent
ResourceTool B
Tool
Tool B
Autonomy
Runs autonomously
Receives
  • writes to ← Agent

How a typical Solo + Tools team handles a task

Typical Solo + Tools layout — schematic, not verified wiring

  1. Task arrives

    Human operator directs Agent.

  2. Agent builds the work

    Agent builds the work.

  3. The artifact lands

    The artifact lands in Tool A: Agent contributes via "writes to". The artifact lands in Tool B: Agent contributes via "writes to".

  4. Human holds the last word

    Human operator holds final approval.

Replicate a typical Solo + Tools setup

Typical Solo + Tools layout — schematic, not verified wiring

Ingredients

  • HumanHuman operator
  • BuilderAgent
  • ResourceTool A
  • ResourceTool B

Setup order

  1. 1.Provision the substrate: Tool A and Tool B.
  2. 2.Wire Agent: it receives "directs" from Human operator.
  3. 3.Declare the human gate: Human operator holds final approval.

Performance metrics

Windowed metrics with provenance. [unknown] means it was not tracked — an honest hole beats an invented figure.

SWE-bench Verified (closest matching run)
64.8%
evidence-linked

Claude 4 Sonnet (20250514) run scoring 64.8% (324/500), 2025-07-26 — the closest actual leaderboard run to swebench.com's rotating 'up to 65%' banner (Wayback snapshot 2025-08-02). The banner's exact model is unstated, and the live banner has since moved past 74%. Source: swebench.com, 2025-08-02 (archived). [evidence_linked]

as of Jul 26, 2025

Token economics

Cost transparency is part of the honesty architecture. [unknown] means it was not tracked — not that it is zero.

No cost metrics on record. Cost tracking is hard across runtimes; honest absence beats invented figures.

Blueprint

Operational DNA — why it works, how it was built, and how it is overseen. Not files for sale; knowledge of the design.

Why it works

Minimalism is the point: by removing scaffold engineering, mini-SWE-agent isolates raw model capability as the variable under test, making it a clean floor/reference point for comparing LMs on SWE-bench.

How it was built

Open source, ~100 lines of Python (SWE-agent/mini-swe-agent on GitHub). Deliberately unengineered so that leaderboard comparisons reflect the underlying model, not scaffold tricks.

Oversight model

None in benchmark evaluation — fully autonomous per-issue runs, intentionally kept minimal to isolate model capability from harness engineering.

Proof (1)

The team's shared track record — tasks, incidents, lessons, milestones. Per-entry provenance tags are always visible.

  1. ArtifactAug 2, 2025evidence-linked

    mini-SWE-agent banner confirmed via archive.org — up to 65% on SWE-bench Verified

    swebench.com's homepage banner ('up to 65%') confirmed via a 2025-08-02 Wayback snapshot; closest matching leaderboard run is Claude 4 Sonnet at 64.8% (324/500), dated 2025-07-26.

    https://www.swebench.com/verified.html

Sign in to add a proof entry.

Sign in

Attestations (0)

Named third-party statements from people with first-hand experience. Attestations are what separates Peer-Attested from Evidence-Linked.

No attestations yet. Worked with this configuration or agent? Attest to it using the form below — attestations are named third-party statements and are what separates Peer-Attested from Evidence-Linked.

Sign in to attest to this team.

Sign in