codework 0.1.0

package/AGENTS.md

@@ -0,0 +1,8 @@
+# Agent Instructions
+
+Codework coordinates AI coding agents through append-only workspace events and stdout-as-context.
+
+MUST: Keep stdout operational and explicit.
+MUST: Preserve the `CODEWORK CONTEXT GUIDE` contract for LLM-facing commands.
+MUST NOT: Add behavior that launches, drives, or controls external AI agent CLIs.
+MUST NOT: add destructive repository automation to Codework without a separate guardrail design.

package/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 SHUHEI IHARA
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

package/README.md

@@ -0,0 +1,55 @@
+# Codework
+
+Codework is a TypeScript CLI harness for coordinating multiple AI coding agents through stdout-as-context.
+
+It does not run Codex, Claude Code, or Cursor CLI. Instead, each agent calls `codework` from its own shell-capable environment, and Codework returns a strict operational guide plus shared workspace state.
+
+## Quick start
+
+In Codex:
+
+```bash
+! codework new workspace=my-workspace --name=codex --follow=user
+```
+
+In Claude Code:
+
+```bash
+! codework join --workspace=my-workspace --name=claude --follow=codex
+```
+
+Then agents should repeatedly use:
+
+```bash
+! codework poll --workspace=my-workspace --name=<agent>
+! codework say --workspace=my-workspace --name=<agent> --to=<agent|all> --kind=note --message="..."
+! codework done --workspace=my-workspace --name=<agent> --summary="..."
+```
+
+## Runtime support
+
+Node.js:
+
+```bash
+npm run build
+node dist/cli.js doctor
+```
+
+Deno:
+
+```bash
+deno run --allow-read --allow-write --allow-env --allow-run=git src/cli.ts doctor
+```
+
+Bun:
+
+```bash
+bun run src/cli.ts doctor
+bun run src/cli.ts new workspace=my-workspace --name=bun --follow=user
+```
+
+## Notes for Japanese users
+
+Codework の stdout は、そのまま LLM の次コンテキストに入る主成果物です。そのため既定の text 出力は短い成功メッセージではなく、エージェントが守るべき英語の運用命令を返します。
+
+`workspace=my-workspace` と `--workspace=my-workspace` の両方を受け付けます。これは Codex / Claude Code などの shell escape 利用時の表記ゆれを吸収するための仕様です。