weave
v0.5.1Weave is a dependency-free engineering layer for Claude Code and Codex. It wraps shell commands to strip noisy successful output while always preserving failures in full, keeps a compact .weave/state.md handoff so context survives across sessions and between agents, and applies one consistent minimalism-and-verification policy so agents reuse existing code, prefer native features over new dependencies, and back every completion claim with real evidence.
By Gabriel CostaLicense: MIT3 GitHub starsUpdated 6 days ago
Directory evidence
- Runtimes
- Codex and Claude Code
- Parsed components
- 6 skill or MCP entries
- Source updated
- Sep 17, 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 weave for Codex and Claude Code
codex plugin marketplace add GabrielKqw/weave
codex plugin marketplace upgrade weave
codex plugin add weave@weavePaste and run these commands in a terminal with Codex. They add and refresh the weave catalog, then install this plugin.
Compatibility: the page URL and API slug “weave-2” remain stable.
- Codex:
weave-2@agent-plugin-marketplace→weave@weave
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/GabrielKqw/weaveClone the source repository, then follow its setup instructions to add the plugin to a compatible client. The repository root is the plugin root.
Plugin files
├── .codex-plugin/plugin.json├── .claude-plugin/plugin.json├── skills/weave/SKILL.md├── skills/weave-audit/SKILL.md├── skills/weave-debt/SKILL.md├── skills/weave-gain/SKILL.md├── skills/weave-help/SKILL.md└── skills/weave-review/SKILL.md
Included Skills6
Use for any coding task that benefits from a request contract, a short pre-edit plan, working memory across a session, and a fidelity check before declaring done - connecting agents, project context, terminal output, tests, git, and prior decisions into one adaptive workflow. Turns a free-form request into a verifiable contract (Goal/Constraints/Required/Not requested/Completion criteria), classifies it as simple or complex, and runs the matching flow. Reads and writes .weave/state.md so Codex and Claude Code can hand work back and forth. Also use when the user says "weave", "use weave", or asks to continue work tracked in .weave/.
Audit an entire repository for removable code, needless dependencies, duplicate helpers, dead flexibility, and custom code replaceable by native platform features. Use when the user asks for a Weave audit, repository simplification audit, or a ranked list of bloat.
Collect deliberate Weave debt markers from a repository into a concise ledger. Use when the user asks for Weave debt, deferred shortcuts, known ceilings, or weave comments.
Show measured terminal-output reduction from local Weave run history. Use when the user asks for Weave gain, token savings, output savings, or the local compression scoreboard.
Explain the available Weave modes, operational skills, and CLI commands. Use when the user asks for Weave help, commands, modes, or how to use the plugin.
Review the current code diff for avoidable complexity, duplicated mechanisms, speculative abstractions, and dependencies that native features or the standard library can replace. Use when the user asks for a Weave review, simplification review, or what can be removed from the current change.
Plugin manifests2
{
"name": "weave",
"version": "0.5.1",
"description": "Weave is a dependency-free engineering layer for Claude Code and Codex. It wraps shell commands to strip noisy successful output while always preserving failures in full, keeps a compact .weave/state.md handoff so context survives across sessions and between agents, and applies one consistent minimalism-and-verification policy so agents reuse existing code, prefer native features over new dependencies, and back every completion claim with real evidence.",
"author": {
"name": "Gabriel Costa",
"url": "https://www.linkedin.com/in/gabriel-costa-940b89276/"
},
"license": "MIT",
"skills": "./skills/",
"hooks": "./hooks/hooks.json",
"interface": {
"displayName": "Weave",
"shortDescription": "Terminal intelligence, context continuity, and verification",
"longDescription": "Wraps shell commands to strip noisy successful output while always preserving failures in full, keeps a compact .weave/state.md handoff across sessions, and applies one minimalism/verification policy.",
"developerName": "Gabriel Costa",
"category": "Productivity",
"capabilities": [
"Instructions",
"Lifecycle hooks"
],
"websiteURL": "https://github.com/GabrielKqw/weave"
}
}{
"name": "weave",
"version": "0.5.1",
"description": "Weave is a dependency-free engineering layer for Claude Code and Codex. It wraps shell commands to strip noisy successful output while always preserving failures in full, keeps a compact .weave/state.md handoff so context survives across sessions and between agents, and applies one consistent minimalism-and-verification policy so agents reuse existing code, prefer native features over new dependencies, and back every completion claim with real evidence.",
"author": {
"name": "Gabriel Costa",
"url": "https://www.linkedin.com/in/gabriel-costa-940b89276/"
},
"license": "MIT",
"keywords": [
"orchestration",
"context",
"workflow",
"terminal",
"verification"
]
}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.
[weave on Agent Plugins Marketplace](https://pluginsmp.com/plugins/weave-2)