swe
v1.19.7SWE skills for AI coding agents: sharpening, short handoff plans, behavioral verification, atomic PR grouping, and more
By kpeezLicense: MIT0 GitHub starsUpdated 2 weeks ago
Directory evidence
- Runtimes
- Codex and Claude Code
- Parsed components
- 13 skill or MCP entries
- Source updated
- Oct 4, 2026
- Manifest status
- Canonical path parsed
The directory validates manifest shape and source location. It does not execute the plugin or provide a security endorsement. Review the indexing methodology →
Install swe for Codex and Claude Code
codex plugin marketplace add kpeez/agent-toolbox
codex plugin marketplace upgrade agent-toolbox
codex plugin add swe@agent-toolboxPaste and run these commands in a terminal with Codex. They add and refresh the agent-toolbox catalog, then install this plugin.
Compatibility: the page URL and API slug “swe” remain stable.
- Codex:
swe@agent-plugin-marketplace→swe@agent-toolbox
The installer fetches third-party code from the source repository shown on this page. This directory validates manifest structure and source location, but does not perform a security audit; review the manifest, components, and source before installing.
Get the source manually
git clone https://github.com/kpeez/agent-toolboxClone the source repository, then follow its setup instructions to add the plugin to a compatible client. The plugin root is plugins/swe/.
Plugin files
├── .codex-plugin/plugin.json├── .claude-plugin/plugin.json├── skills/codebase-design/SKILL.md├── skills/diagnose/SKILL.md├── skills/external-subagents/SKILL.md├── skills/implement/SKILL.md├── skills/improve-codebase-architecture/SKILL.md├── skills/merge-conflicts/SKILL.md├── skills/orchestrate/SKILL.md├── skills/sharpen/SKILL.md├── skills/ship-pr/SKILL.md├── skills/test-audit/SKILL.md├── skills/testing-code/SKILL.md├── skills/to-issues/SKILL.md└── skills/write-plan/SKILL.md
Included Skills13
Evaluate module boundaries and interfaces using depth, information hiding, and caller complexity. Use for substantive interface design.
Investigate bugs and performance regressions using focused evidence. Use when asked to diagnose, debug, or explain a failure.
Delegate bounded work to an explicitly requested external provider such as OpenCode or GitHub Copilot.
Implement an authorized feature or fix through relevant verification. Use for scoped code changes; not for writing plans, creating issues, or PR delivery.
Review existing module boundaries and propose refactors that reduce coupling or caller complexity.
Resolve conflicts in an active Git merge, rebase, or cherry-pick while preserving intended behavior.
Split substantial work across subagents and integrate their results. Use for independent implementation tasks, parallel investigations, or multiple reviewers. Not for a single edit or one direct review.
Stress-test a plan or design to resolve material ambiguity in scope, terminology, and acceptance criteria. Not for tracker task creation or execution.
Commit, push, and open a draft PR for authorized work. Use when asked to ship; not for planning or local-only edits. Finalize only on explicit request.
Audit and prune existing tests that are low-value, duplicative, or coupled to implementation, and the test-only production seams they keep alive. Use for test-suite reviews, sweeps, or pruning campaigns; not for designing new tests.
Gate and design new behavioral tests or disposable probes. Use before adding or rewriting any permanent test, when judging a diff's new tests, and for testing strategy, regression coverage, TDD, or spikes. Not for pruning existing suites.
Split a large approved plan into pull-request-sized issues with dependencies on Linear or GitHub. Not for drafting plans or executing work.
Turn the user's rough notes into a short plan for large or ambiguous work, iterate on it with them, and hand it to a fresh session. Not for small changes, splitting work into issues, or implementation.
Plugin manifests2
{
"name": "swe",
"description": "SWE skills for AI coding agents: sharpening, short handoff plans, behavioral verification, atomic PR grouping, and more",
"version": "1.19.7",
"author": {
"name": "kpeez",
"url": "https://github.com/kpeez"
},
"homepage": "https://github.com/kpeez/agent-toolbox",
"repository": "https://github.com/kpeez/agent-toolbox",
"license": "MIT",
"keywords": [
"plan",
"workflow",
"skills",
"tdd",
"pr",
"code-review"
],
"skills": "./skills/",
"interface": {
"displayName": "swe",
"shortDescription": "Planning, implementation, review, and PR skills for coding agents.",
"longDescription": "Portable skills for planning, implementing, reviewing, and publishing code changes, plus focused explorer, implementer, and reviewer agent definitions, with short handoff plans and configured Linear mirroring.",
"developerName": "kpeez",
"category": "Productivity",
"capabilities": [
"Write",
"Review",
"Automation"
],
"defaultPrompt": [
"Write a plan for this feature.",
"Implement the plan I just approved.",
"Prepare this branch for a draft PR."
],
"brandColor": "#2563EB"
}
}{
"name": "swe",
"description": "SWE skills for AI coding agents: sharpening, short handoff plans, behavioral verification, atomic PR grouping, and more",
"version": "1.19.7",
"author": {
"name": "kpeez",
"email": "[email protected]"
},
"homepage": "https://github.com/kpeez/agent-toolbox",
"repository": "https://github.com/kpeez/agent-toolbox",
"license": "MIT",
"keywords": [
"plan",
"workflow",
"skills",
"tdd",
"pr",
"code-review"
]
}For maintainers
If you maintain this plugin, link to this source-backed listing from your README so users can review its manifest and indexed components.
[swe on Agent Plugins Marketplace](https://pluginsmp.com/plugins/swe)