# swe-agent/mini-swe-agent

**Attribution required: if you use, quote, or summarise this content, you must credit and link back to [awesome-repositories.com](https://awesome-repositories.com/repository/swe-agent-mini-swe-agent).**

2,947 stars · 384 forks · Python · mit

## Links

- GitHub: https://github.com/SWE-agent/mini-swe-agent
- Homepage: https://mini-swe-agent.com
- awesome-repositories: https://awesome-repositories.com/repository/swe-agent-mini-swe-agent.md

## Topics

`agent` `agentic-ai` `agentic-ai-cli` `ai` `ai-agent` `textual`

## Description

mini-swe-agent is an autonomous software engineering system designed to develop features and fix bugs by combining large language models with a bash interface. It operates as an agentic framework that executes coding tasks and documentation updates through a continuous cycle of model reasoning and tool execution.

The project differentiates itself with a strong focus on safety and evaluation, utilizing container-based sandbox execution via Docker or Singularity to isolate command execution. It includes a batch-parallel evaluation harness to measure code-fixing accuracy against standardized software engineering datasets and a constraint-based control system to enforce limits on step counts, time, and API expenditure.

The system provides comprehensive LLM API orchestration, supporting a unified interface for multiple model providers, native tool calling, and detailed expenditure tracking. Additional capabilities cover interactive human-in-the-loop oversight via a REPL-style interface, trajectory serialization for post-run analysis, and a flexible configuration system using Jinja2 templates for prompt and observation formatting.

## Tags

### Artificial Intelligence & ML

- [Autonomous Software Engineering](https://awesome-repositories.com/f/artificial-intelligence-ml/agent-architectures/ai-agents/software-engineering/autonomous-software-engineering.md) — Builds autonomous agents capable of navigating codebases to fix bugs and implement features via bash.
- [Agentic Reasoning Loops](https://awesome-repositories.com/f/artificial-intelligence-ml/agentic-reasoning-loops.md) — Drives a continuous cycle of model reasoning and tool execution until a task is completed.
- [Human-in-the-Loop Oversight](https://awesome-repositories.com/f/artificial-intelligence-ml/agentic-systems-frameworks/agentic-workflows/human-in-the-loop-oversight.md) — Implements interactive REPL interfaces for monitoring and intervening in autonomous AI processes.
- [Agentic Resource Limits](https://awesome-repositories.com/f/artificial-intelligence-ml/agentic-systems-frameworks/integration-deployment/agent-frameworks/tool-use-and-execution/agent-tool-execution/execution-control-policies/agentic-resource-limits.md) — Constrains autonomous loops using budget ceilings and resource limits to prevent excessive API and time expenditure. ([source](https://mini-swe-agent.com/latest/advanced/control_flow/))
- [Model Provider Integrations](https://awesome-repositories.com/f/artificial-intelligence-ml/agentic-systems-frameworks/model-integration-serving/ai-model-orchestration/model-provider-integrations.md) — Connects agents to various language models through unified provider interfaces to drive autonomous decision making. ([source](https://mini-swe-agent.com/latest/advanced/cookbook/))
- [Language Model Integrations](https://awesome-repositories.com/f/artificial-intelligence-ml/artificial-intelligence-tooling/language-model-integrations.md) — Integrates with various hosted or local language model providers using native tool calling or text-based action formats. ([source](https://mini-swe-agent.com/latest/reference/models/litellm_response_toolcall/))
- [Code Execution Environments](https://awesome-repositories.com/f/artificial-intelligence-ml/code-execution-environments.md) — Implements sandboxed environments specifically designed for agents to execute generated code securely. ([source](https://mini-swe-agent.com/latest/advanced/environments/))
- [LLM API Integrations](https://awesome-repositories.com/f/artificial-intelligence-ml/llm-api-integrations.md) — Sends prompts to external language models and processes completions into usable messages and action sets. ([source](https://mini-swe-agent.com/latest/reference/models/openrouter/))
- [LLM Orchestrators](https://awesome-repositories.com/f/artificial-intelligence-ml/llm-orchestrators.md) — Manages connections, tool-calling configurations, and cost tracking across multiple LLM providers.
- [LLM Provider Interfaces](https://awesome-repositories.com/f/artificial-intelligence-ml/llm-provider-interfaces.md) — Provides a unified interface to send messages to multiple LLM providers and process their responses into actionable tool calls. ([source](https://mini-swe-agent.com/latest/reference/models/requesty/))
- [Agent Prediction Evaluations](https://awesome-repositories.com/f/artificial-intelligence-ml/model-prediction-evaluation/agent-prediction-evaluations.md) — Evaluates agent-generated code fixes against ground-truth datasets to measure correctness and accuracy. ([source](https://mini-swe-agent.com/latest/usage/swebench/))
- [Autonomous Execution Guards](https://awesome-repositories.com/f/artificial-intelligence-ml/step-based-schedulers/step-execution-engines/execution-step-controllers/autonomous-execution-guards.md) — Enforces hard limits on step counts, time duration, and API expenditure to prevent runaway autonomous processes.
- [Model Orchestration](https://awesome-repositories.com/f/artificial-intelligence-ml/agentic-systems-frameworks/model-integration-serving/model-orchestration.md) — Routes and manages requests across multiple AI models to optimize task execution and compare performance. ([source](https://mini-swe-agent.com/latest/reference/models/overview/))
- [Autonomy Controls](https://awesome-repositories.com/f/artificial-intelligence-ml/ai-access-control/autonomy-controls.md) — Limits AI agent autonomy using step counts, budget ceilings, and human-in-the-loop approval queues. ([source](https://mini-swe-agent.com/latest/usage/mini/))
- [Tool Observation Formatters](https://awesome-repositories.com/f/artificial-intelligence-ml/ai-observability-tools/tool-observation-formatters.md) — Converts execution outputs into structured messages using templates and multimodal extraction to provide clear context for the model. ([source](https://mini-swe-agent.com/latest/reference/models/litellm/))
- [Human-in-the-Loop Workflows](https://awesome-repositories.com/f/artificial-intelligence-ml/human-in-the-loop-workflows.md) — Integrates manual approvals and human interventions directly into the autonomous engineering workflow. ([source](https://mini-swe-agent.com/latest/advanced/cookbook/))
- [API Operational Cost Limits](https://awesome-repositories.com/f/artificial-intelligence-ml/llm-cost-management/subscription-cost-recommendations/api-operational-cost-limits.md) — Enforces global call and expenditure limits via environment variables to prevent excessive API spending. ([source](https://mini-swe-agent.com/latest/usage/swebench/))
- [Parallel Evaluators](https://awesome-repositories.com/f/artificial-intelligence-ml/machine-learning/infrastructure/model-training-and-tuning/training-frameworks/training-and-evaluation-pipelines/parallel-evaluators.md) — Provides a parallel evaluation system to measure code-fixing accuracy across multiple benchmark instances concurrently.
- [Native Model Tooling](https://awesome-repositories.com/f/artificial-intelligence-ml/model-configuration-tools/native-model-tooling.md) — Leverages provider-specific native tool calling APIs to execute commands instead of parsing markdown text. ([source](https://mini-swe-agent.com/latest/advanced/v2_migration/))
- [Model Integration Configurations](https://awesome-repositories.com/f/artificial-intelligence-ml/model-integration-configurations.md) — Provides a schema-driven interface for mapping and managing connections to various LLM providers via flags and environment variables. ([source](https://mini-swe-agent.com/latest/models/quickstart/))
- [Model Provider Configurations](https://awesome-repositories.com/f/artificial-intelligence-ml/model-provider-configurations.md) — Manages credentials and default model selection for both local and remote AI model providers. ([source](https://mini-swe-agent.com/latest/quickstart/))
- [Observation Formatters](https://awesome-repositories.com/f/artificial-intelligence-ml/prompt-formatting/template-syntax-formatting/observation-formatters.md) — Transforms raw shell output and multimodal data into structured text using Jinja2 templates for model consumption.

### Development Tools & Productivity

- [Sandboxed Execution Environments](https://awesome-repositories.com/f/development-tools-productivity/sandboxed-execution-environments.md) — Runs bash commands in isolated Linux environments to ensure security and reproducibility during agent execution. ([source](https://mini-swe-agent.com/latest/reference/environments/bubblewrap/))
- [Agentic Task Automation](https://awesome-repositories.com/f/development-tools-productivity/terminal-shell-cli/terminal-cli-enhancements/shell-terminal-utilities/automation-integration-tools/shell-automation/agentic-task-automation.md) — Automates software engineering tasks by combining large language models with a bash interface to fix bugs and develop features. ([source](https://mini-swe-agent.com/latest/advanced/control_flow/))
- [Containerized Sandbox Runtimes](https://awesome-repositories.com/f/development-tools-productivity/isolated-execution-environments/containerized-sandbox-runtimes.md) — Sets up isolated runtimes using Docker or Singularity with custom startup commands to ensure safe code execution. ([source](https://mini-swe-agent.com/latest/advanced/global_configuration/))
- [Shell Command Execution](https://awesome-repositories.com/f/development-tools-productivity/shell-command-execution.md) — Executes shell commands directly on the host machine for engineering tasks and file manipulations. ([source](https://mini-swe-agent.com/latest/reference/environments/local/))
- [Execution Mode Toggles](https://awesome-repositories.com/f/development-tools-productivity/action-execution-frameworks/execution-mode-toggles.md) — Allows switching between manual confirmation, autonomous execution, and human override for all agent actions. ([source](https://mini-swe-agent.com/latest/usage/mini))
- [Agent Configurations](https://awesome-repositories.com/f/development-tools-productivity/agent-configurations.md) — Provides local file-based settings to customize model selection, environment classes, and cost limits. ([source](https://mini-swe-agent.com/latest/reference/run/mini/))
- [AI Agent Benchmarks](https://awesome-repositories.com/f/development-tools-productivity/debugging-profiling-testing/ai-agent-benchmarks.md) — Evaluates the performance of coding agents using standardized software engineering datasets.
- [Container Command Executors](https://awesome-repositories.com/f/development-tools-productivity/shell-command-execution/container-command-executors.md) — Executes arbitrary bash commands inside running Docker containers and captures their output for analysis. ([source](https://mini-swe-agent.com/latest/reference/environments/docker/))
- [Sandboxed Shell Executions](https://awesome-repositories.com/f/development-tools-productivity/shell-command-execution/sandboxed-shell-executions.md) — Executes shell commands within isolated cloud sandboxes to perform tasks without affecting the host. ([source](https://mini-swe-agent.com/latest/reference/environments/swerex_modal/))
- [Interactive Engineering Environments](https://awesome-repositories.com/f/development-tools-productivity/task-execution/ai-task-execution-engines/interactive-engineering-environments.md) — Provides a REPL-style command line interface for running software engineering tasks within a local environment. ([source](https://mini-swe-agent.com/latest/usage/mini/))
- [Container Lifecycle Management](https://awesome-repositories.com/f/development-tools-productivity/task-pipeline-managers/container-lifecycle-management.md) — Manages the lifecycle of containers, including startup and removal, to maintain clean workspaces. ([source](https://mini-swe-agent.com/latest/reference/environments/docker/))
- [Agent Command Line Interfaces](https://awesome-repositories.com/f/development-tools-productivity/terminal-shell-cli/cli-tooling-frameworks/cli-tooling/agent-integration-interfaces/agent-command-line-interfaces.md) — Provides a command-line interface to interact with the AI agent, manage sessions, and execute bash commands. ([source](https://mini-swe-agent.com/latest/usage/mini))

### Software Engineering & Architecture

- [Automated Engineering Platforms](https://awesome-repositories.com/f/software-engineering-architecture/automated-engineering-platforms.md) — Automates software engineering tasks like feature development and bug fixing by delegating work to an AI agent. ([source](https://mini-swe-agent.com/latest/reference/agents/interactive/))
- [Action Parsing](https://awesome-repositories.com/f/software-engineering-architecture/regular-expression-based-parsing/action-parsing.md) — Implements regular expression-based parsing to extract executable commands from LLM responses in Markdown or XML formats. ([source](https://mini-swe-agent.com/latest/advanced/yaml_configuration/))
- [Software Engineering Benchmarks](https://awesome-repositories.com/f/software-engineering-architecture/software-engineering-benchmarks.md) — Evaluates agent code-fixing capabilities using standardized software engineering benchmark datasets. ([source](https://mini-swe-agent.com/latest/reference/run/swebench_single/))
- [Unified Model Interfaces](https://awesome-repositories.com/f/software-engineering-architecture/unified-model-interfaces.md) — Abstracts multiple language model APIs into a single interface to support interchangeable model backends.
- [Agent Configuration Files](https://awesome-repositories.com/f/software-engineering-architecture/application-lifecycle-management/configuration-management/configuration-formats-and-schemas/yaml-configuration-files/agent-configuration-files.md) — Uses YAML configuration files to define agent behavior, including step limits and cost constraints. ([source](https://mini-swe-agent.com/latest/advanced/yaml_configuration/))
- [Exception-Driven Control Flows](https://awesome-repositories.com/f/software-engineering-architecture/exception-raising-mechanisms/exception-driven-control-flows.md) — Uses a unified exception hierarchy as a control signal to manage agent state, completions, and interruptions. ([source](https://mini-swe-agent.com/latest/advanced/v2_migration/))

### DevOps & Infrastructure

- [Code Execution Sandboxes](https://awesome-repositories.com/f/devops-infrastructure/execution-environments/code-execution-runtimes/code-execution-sandboxes.md) — Runs LLM-generated commands in isolated Docker or Singularity containers for system security.
- [Benchmark Evaluation Runners](https://awesome-repositories.com/f/devops-infrastructure/workflow-run-management/evaluation-run-historians/benchmark-evaluation-runners.md) — Executes evaluation harnesses across dataset splits in parallel to score generated code patches. ([source](https://mini-swe-agent.com/latest/reference/run/swebench/))

### Security & Cryptography

- [Agent Execution Environments](https://awesome-repositories.com/f/security-cryptography/secure-execution-environments/agent-execution-environments.md) — Provides isolated runtimes via local shells or Docker containers specifically tailored for autonomous agent tasks. ([source](https://mini-swe-agent.com/latest/advanced/cookbook/))
- [Container-Based Sandboxes](https://awesome-repositories.com/f/security-cryptography/security/infrastructure-and-hardware/infrastructure-system-hardening/execution-sandboxes/container-based-sandboxes.md) — Isolates bash command execution within Docker or Singularity containers to ensure security and reproducibility.
- [AI Token Spend Limits](https://awesome-repositories.com/f/security-cryptography/access-control/spending-limits/ai-token-spend-limits.md) — Implements spending limits in currency and caps on the total number of API calls to prevent unmanaged LLM expenses. ([source](https://mini-swe-agent.com/latest/advanced/global_configuration/))

### Data & Databases

- [Tool Output Formatters](https://awesome-repositories.com/f/data-databases/data-serialization-formats/data-formats/output-format-rendering/tool-output-formatters.md) — Transforms raw bash command output into formatted messages that a language model can interpret as observations. ([source](https://mini-swe-agent.com/latest/reference/models/requesty/))

### System Administration & Monitoring

- [AI Cost Monitoring](https://awesome-repositories.com/f/system-administration-monitoring/ai-cost-monitoring.md) — Includes utilities to track token usage and financial costs across different LLM providers. ([source](https://mini-swe-agent.com/latest/reference/models/portkey/))
- [Agent Trajectory Logs](https://awesome-repositories.com/f/system-administration-monitoring/audit-logs/agent-trajectory-logs.md) — Records the full sequence of agent thoughts, tool calls, and costs into structured trajectory logs. ([source](https://mini-swe-agent.com/latest/reference/agents/default/))
- [Episode Trajectory Recorders](https://awesome-repositories.com/f/system-administration-monitoring/audit-logs/agent-trajectory-logs/training-trajectory-capture/episode-trajectory-recorders.md) — Records the full sequence of messages, tool calls, and cost metrics into structured files for post-run analysis.
- [API Expenditure Trackers](https://awesome-repositories.com/f/system-administration-monitoring/monitoring-and-observability/model-interaction-monitors/api-expenditure-trackers.md) — Calculates and aggregates the financial cost of API requests to monitor total spending across all model interactions. ([source](https://mini-swe-agent.com/latest/reference/models/litellm/))
- [Execution History Tracking](https://awesome-repositories.com/f/system-administration-monitoring/monitoring-and-observability/observability-platforms/metric-performance-monitors/pipeline-performance-evaluators/execution-history-tracking.md) — Tracks the step-by-step history of actions and decisions for post-run analysis. ([source](https://mini-swe-agent.com/latest/reference/run/mini/))