pilot-workers 0.2.0__tar.gz

pilot_workers-0.2.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026
+
+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.

pilot_workers-0.2.0/PKG-INFO

@@ -0,0 +1,84 @@
+Metadata-Version: 2.4
+Name: pilot-workers
+Version: 0.2.0
+Summary: Dispatch bounded tasks to isolated LLM workers (GLM, Kimi, and more)
+License-Expression: MIT
+Requires-Python: >=3.11
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Provides-Extra: yaml
+Requires-Dist: pyyaml>=6.0; extra == "yaml"
+Provides-Extra: dev
+Requires-Dist: pytest>=8.0; extra == "dev"
+Requires-Dist: pyyaml>=6.0; extra == "dev"
+Dynamic: license-file
+
+# pilot-workers
+
+Dispatch bounded tasks to isolated LLM workers. Your main AI agent (Claude, Codex, or any planner) stays in control of requirements, planning, and verification — the worker only executes what it's told.
+
+## What it does
+
+- **Provider isolation**: each model (GLM, Kimi, or your own) gets its own credentials, XDG directories, logs, and session storage. No cross-contamination.
+- **Fixed routing**: provider, model, and endpoint are locked per YAML config. Tasks cannot override them.
+- **Security by default**: API keys never appear in CLI args, environment variables, task contracts, or logs. Output is auto-redacted.
+- **Five modes**: `code` (edit), `explore` (read-only), `test` (run tests), `review` (read-only audit), `resume` (continue a prior code session).
+- **Observable**: `worker_runner.started` → heartbeats → `worker_runner.summary`. Human-readable live log at `latest.log` for `tail -f`.
+- **Pluggable runners**: currently wraps [OpenCode](https://opencode.ai) via `@ai-sdk/openai-compatible`; the runner interface is designed for future alternatives (Aider, Continue, etc.).
+
+## Quick start
+
+```bash
+# 1. Install the pinned OpenCode runtime
+pilot-workers runtime install
+
+# 2. Add a provider (YAML files in src/pilot_workers/data/providers/)
+# GLM and Kimi are included; add your own by copying the template
+
+# 3. Configure credentials (interactive, key never displayed)
+python3 -m pilot_workers.credentials glm
+python3 -m pilot_workers.credentials kimi-k3
+
+# 4. Verify
+python3 -m pilot_workers.cli.run --provider glm --mode explore --workdir . --task "hello" --dry-run
+
+# 5. Run a real task
+python3 -m pilot_workers.cli.run \
+  --provider glm --mode code \
+  --workdir /path/to/project \
+  --task-file /path/to/task.md
+```
+
+## Adding a new provider
+
+Drop a YAML file in `data/providers/` (inside the package):
+
+```yaml
+key: deepseek
+provider_id: deepseek-worker
+model_id: deepseek-coder-v3
+base_url: https://api.deepseek.com/v1
+display_name: DeepSeek Coder V3
+context_tokens: 128000
+output_tokens: 8192
+```
+
+Then `python3 -m pilot_workers.credentials deepseek` and you're done.
+
+## Host integration
+
+The **host** is whatever AI agent acts as the planner — Claude, Codex, Gemini, GLM itself, or anything that can write a task file and call the CLI.
+
+`integrations/` has ready-made configs for current hosts:
+- **`claude-host/`**: 12 agents + 8 slash commands (`/glm:code`, `/kimi:explore`, etc.)
+- **`codex-host/`**: `$glm` / `$kimi` / `$ds` skill entry points
+
+Adding a new host: create `integrations/<name>-host/`, put whatever config your host needs, point it at `python3 -m pilot_workers.cli.run`. See `integrations/README.md`.
+
+## Architecture
+
+See [docs/architecture.md](docs/architecture.md) for the shared contract, security model, and verification checklist.
+
+## License
+
+MIT

pilot_workers-0.2.0/README.md

@@ -0,0 +1,69 @@
+# pilot-workers
+
+Dispatch bounded tasks to isolated LLM workers. Your main AI agent (Claude, Codex, or any planner) stays in control of requirements, planning, and verification — the worker only executes what it's told.
+
+## What it does
+
+- **Provider isolation**: each model (GLM, Kimi, or your own) gets its own credentials, XDG directories, logs, and session storage. No cross-contamination.
+- **Fixed routing**: provider, model, and endpoint are locked per YAML config. Tasks cannot override them.
+- **Security by default**: API keys never appear in CLI args, environment variables, task contracts, or logs. Output is auto-redacted.
+- **Five modes**: `code` (edit), `explore` (read-only), `test` (run tests), `review` (read-only audit), `resume` (continue a prior code session).
+- **Observable**: `worker_runner.started` → heartbeats → `worker_runner.summary`. Human-readable live log at `latest.log` for `tail -f`.
+- **Pluggable runners**: currently wraps [OpenCode](https://opencode.ai) via `@ai-sdk/openai-compatible`; the runner interface is designed for future alternatives (Aider, Continue, etc.).
+
+## Quick start
+
+```bash
+# 1. Install the pinned OpenCode runtime
+pilot-workers runtime install
+
+# 2. Add a provider (YAML files in src/pilot_workers/data/providers/)
+# GLM and Kimi are included; add your own by copying the template
+
+# 3. Configure credentials (interactive, key never displayed)
+python3 -m pilot_workers.credentials glm
+python3 -m pilot_workers.credentials kimi-k3
+
+# 4. Verify
+python3 -m pilot_workers.cli.run --provider glm --mode explore --workdir . --task "hello" --dry-run
+
+# 5. Run a real task
+python3 -m pilot_workers.cli.run \
+  --provider glm --mode code \
+  --workdir /path/to/project \
+  --task-file /path/to/task.md
+```
+
+## Adding a new provider
+
+Drop a YAML file in `data/providers/` (inside the package):
+
+```yaml
+key: deepseek
+provider_id: deepseek-worker
+model_id: deepseek-coder-v3
+base_url: https://api.deepseek.com/v1
+display_name: DeepSeek Coder V3
+context_tokens: 128000
+output_tokens: 8192
+```
+
+Then `python3 -m pilot_workers.credentials deepseek` and you're done.
+
+## Host integration
+
+The **host** is whatever AI agent acts as the planner — Claude, Codex, Gemini, GLM itself, or anything that can write a task file and call the CLI.
+
+`integrations/` has ready-made configs for current hosts:
+- **`claude-host/`**: 12 agents + 8 slash commands (`/glm:code`, `/kimi:explore`, etc.)
+- **`codex-host/`**: `$glm` / `$kimi` / `$ds` skill entry points
+
+Adding a new host: create `integrations/<name>-host/`, put whatever config your host needs, point it at `python3 -m pilot_workers.cli.run`. See `integrations/README.md`.
+
+## Architecture
+
+See [docs/architecture.md](docs/architecture.md) for the shared contract, security model, and verification checklist.
+
+## License
+
+MIT

pilot_workers-0.2.0/pyproject.toml

@@ -0,0 +1,39 @@
+[build-system]
+requires = ["setuptools>=68.0"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "pilot-workers"
+version = "0.2.0"
+description = "Dispatch bounded tasks to isolated LLM workers (GLM, Kimi, and more)"
+readme = "README.md"
+license = "MIT"
+requires-python = ">=3.11"
+dependencies = []
+
+[project.optional-dependencies]
+yaml = ["pyyaml>=6.0"]
+dev = ["pytest>=8.0", "pyyaml>=6.0"]
+
+[project.scripts]
+pilot-workers = "pilot_workers.cli.main:main"
+
+[tool.setuptools.packages.find]
+where = ["src"]
+
+[tool.setuptools.package-data]
+pilot_workers = [
+    "data/providers/*.yaml",
+    "data/providers/*.md",
+    "data/permissions/*.yaml",
+    "data/permissions/*.md",
+    "data/templates/*.md",
+    "prompts/*.md",
+    "integrations/**/*.md",
+    "integrations/**/*.yaml",
+    "scripts/*.sh",
+]
+
+[tool.pytest.ini_options]
+testpaths = ["tests"]
+pythonpath = ["src"]

pilot_workers-0.2.0/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

pilot_workers-0.2.0/src/pilot_workers/__init__.py

@@ -0,0 +1 @@
+"""pilot-workers: dispatch bounded tasks to isolated LLM workers."""

pilot_workers-0.2.0/src/pilot_workers/__main__.py

@@ -0,0 +1,3 @@
+from pilot_workers.cli.main import main
+
+raise SystemExit(main())