@kindgi/cli 0.1.4-rc.2 → 0.1.4-rc.4

package/dist/templates/minimal/README.md.tmpl

@@ -48,9 +48,11 @@ kindgi runs start --agent=demo.echo-agent --input='{"userMessage":"hi"}'
 
 ## Use a real model
 
-`kindgi dev` gives a new pack `dev-echo`, a fallback provider that calls the
-first tool and replies "Tool responded: …" — it answers only while no other
-provider fits. To use Claude, with `kindgi dev` running, from another terminal:
+`kindgi dev` gives a new pack `dev-echo`, a fallback provider that isn't a
+model: it calls the first tool and replies "Tool responded: …", after a
+warning line that says so. It answers only while no other provider fits. To
+use a real model, set one LLM provider's key and register its preset, with
+`kindgi dev` running, from another terminal. For Claude:
 
 ```sh
 kindgi secrets set ANTHROPIC_API_KEY --env=local --scope=tenant   # no-echo prompt; writes .env.local
@@ -58,8 +60,12 @@ kindgi providers register --preset=anthropic                      # Opus 5.5, So
 ```
 
 Claude takes over from dev-echo on the next turn; nothing to restart.
-`--models=claude-haiku-4-5` registers one model only; `kindgi providers
-presets` lists the presets (Gemini on Vertex AI too). Picking a model per
+`--models=claude-haiku-4-5` registers one model only. OpenAI
+(`OPENAI_API_KEY`, `--preset=openai`), Gemini (`GEMINI_API_KEY`,
+`--preset=gemini-api`), Groq (`GROQ_API_KEY`, `--preset=groq`) and
+OpenRouter, many vendors' models with one key (`OPENROUTER_API_KEY`,
+`--preset=openrouter`), work the same way; `kindgi providers presets` lists
+the presets (Gemini on Vertex AI too). Picking a model per
 agent, and other providers: `.claude/skills/kindgi-authoring-providers/SKILL.md`.
 
 ## Note on `@kindgi/sdk` versioning

package/dist/templates/minimal/package.json.tmpl

@@ -21,5 +21,8 @@
     "@types/node": "^22.0.0",
     "typescript": "^5.7.3",
     "vitest": "^3.0.0"
+  },
+  "allowScripts": {
+    "esbuild": false
   }
 }

package/dist/templates/minimal/pnpm-workspace.yaml

@@ -3,6 +3,7 @@
 # its install script (true) or skips it (false); pnpm 11+ stops an install
 # until each one has a decision. esbuild (the CLI's bundler, and vitest's
 # transform) works without its script: its native binary comes from its
-# `@esbuild/<platform>` package, so the script stays off.
+# `@esbuild/<platform>` package, so the script stays off. npm 11+ records
+# the same decision in `package.json`'s `allowScripts`.
 allowBuilds:
   esbuild: false

package/dist/templates/python/AGENTS.md

@@ -11,10 +11,13 @@ starts with `_`. The config is `[tool.kindgi]` in `pyproject.toml`.
 - `uv run pytest` runs the tests; a `Tool` or `Guardrail` is still callable.
 - `uv run python -m kindgi.pack index --pack-dir .` shows what Kindgi sees.
 - Agents answer through a model provider. `kindgi dev` gives a new pack
-  `dev-echo`, a fallback that calls the first tool and replies
-  "Tool responded: …" while no other provider fits. For Claude,
-  put the key in `.env` and add a `[[tool.kindgi.providers]]` table with
-  `preset = "anthropic"` to `pyproject.toml`: `kindgi dev` then registers it
+  `dev-echo`, a fallback that isn't a model: it calls the first tool and
+  replies "Tool responded: …" after a warning line, while no other provider
+  fits. For a real model, put one LLM provider's key in `.env`
+  (`ANTHROPIC_API_KEY`, `OPENAI_API_KEY`, `GEMINI_API_KEY`, `GROQ_API_KEY` or
+  `OPENROUTER_API_KEY`) and add a `[[tool.kindgi.providers]]` table with its
+  preset (`preset = "anthropic"`, `"openai"`, `"gemini-api"`, `"groq"` or
+  `"openrouter"`) to `pyproject.toml`: `kindgi dev` then registers it
   on every boot, in every worktree. Other providers and per-agent model
   choice: `.claude/skills/kindgi-authoring-providers/SKILL.md`.
 

package/dist/templates/python/README.md.tmpl

@@ -35,9 +35,11 @@ kindgi runs start --agent={{PACK_ID}}.echo-agent --input='{"userMessage":"hi"}'
 
 ## Use a real model
 
-`kindgi dev` gives a new pack `dev-echo`, a fallback provider that calls the
-first tool and replies "Tool responded: …" — it answers only while no other
-provider fits. To use Claude, with `kindgi dev` running, from another terminal:
+`kindgi dev` gives a new pack `dev-echo`, a fallback provider that isn't a
+model: it calls the first tool and replies "Tool responded: …", after a
+warning line that says so. It answers only while no other provider fits. To
+use a real model, set one LLM provider's key and register its preset, with
+`kindgi dev` running, from another terminal. For Claude:
 
 ```sh
 kindgi secrets set ANTHROPIC_API_KEY --env=local --scope=tenant   # no-echo prompt; writes .env.local
@@ -45,8 +47,12 @@ kindgi providers register --preset=anthropic                      # Opus 5.5, So
 ```
 
 Claude takes over from dev-echo on the next turn; nothing to restart.
-`--models=claude-haiku-4-5` registers one model only; `kindgi providers
-presets` lists the presets (Gemini on Vertex AI too). Picking a model per
+`--models=claude-haiku-4-5` registers one model only. OpenAI
+(`OPENAI_API_KEY`, `--preset=openai`), Gemini (`GEMINI_API_KEY`,
+`--preset=gemini-api`), Groq (`GROQ_API_KEY`, `--preset=groq`) and
+OpenRouter, many vendors' models with one key (`OPENROUTER_API_KEY`,
+`--preset=openrouter`), work the same way; `kindgi providers presets` lists
+the presets (Gemini on Vertex AI too). Picking a model per
 agent, and other providers: `.claude/skills/kindgi-authoring-providers/SKILL.md`.
 
 ## Build

package/dist/templates/python/pyproject.toml.tmpl

@@ -6,7 +6,7 @@ requires-python = ">=3.11"
 dependencies = [{{KINDGI_REQUIREMENT}}]
 
 [dependency-groups]
-dev = ["pytest>=8"]
+dev = [{{DEV_DEPENDENCIES}}]
 
 # The uv versions this pack works with: `kindgi build` installs it with the
 # uv in Kindgi's image, which reads the uv.lock they write.

package/dist/templates/sample/AGENTS.md

@@ -7,9 +7,12 @@ field-level docs. Iterate with `kindgi dev`; `kindgi --help` lists
 the full CLI.
 
 Agents answer through a model provider. `kindgi dev` gives a new pack
-`dev-echo`, a fallback that calls the first tool and replies
-"Tool responded: …" while no other provider fits. For Claude,
-put the key in `.env` and declare `{ preset: 'anthropic' }` in
+`dev-echo`, a fallback that isn't a model: it calls the first tool and
+replies "Tool responded: …" after a warning line, while no other provider
+fits. For a real model, put one LLM provider's key in `.env`
+(`ANTHROPIC_API_KEY`, `OPENAI_API_KEY`, `GEMINI_API_KEY`, `GROQ_API_KEY` or
+`OPENROUTER_API_KEY`) and declare its preset (`{ preset: 'anthropic' }`,
+`'openai'`, `'gemini-api'`, `'groq'` or `'openrouter'`) in
 `kindgi.config.ts`'s `providers`: `kindgi dev` then registers it on every
 boot, in every worktree. Other providers and per-agent model choice:
 `.claude/skills/kindgi-authoring-providers/SKILL.md`.

package/dist/templates/sample/README.md.tmpl

@@ -40,9 +40,11 @@ re-registers on every file save.
 
 ## Use a real model
 
-`kindgi dev` gives a new pack `dev-echo`, a fallback provider that calls the
-first tool and replies "Tool responded: …" — it answers only while no other
-provider fits. To use Claude, with `kindgi dev` running, from another terminal:
+`kindgi dev` gives a new pack `dev-echo`, a fallback provider that isn't a
+model: it calls the first tool and replies "Tool responded: …", after a
+warning line that says so. It answers only while no other provider fits. To
+use a real model, set one LLM provider's key and register its preset, with
+`kindgi dev` running, from another terminal. For Claude:
 
 ```sh
 kindgi secrets set ANTHROPIC_API_KEY --env=local --scope=tenant   # no-echo prompt; writes .env.local
@@ -50,8 +52,12 @@ kindgi providers register --preset=anthropic                      # Opus 5.5, So
 ```
 
 Claude takes over from dev-echo on the next turn; nothing to restart.
-`--models=claude-haiku-4-5` registers one model only; `kindgi providers
-presets` lists the presets (Gemini on Vertex AI too). Picking a model per
+`--models=claude-haiku-4-5` registers one model only. OpenAI
+(`OPENAI_API_KEY`, `--preset=openai`), Gemini (`GEMINI_API_KEY`,
+`--preset=gemini-api`), Groq (`GROQ_API_KEY`, `--preset=groq`) and
+OpenRouter, many vendors' models with one key (`OPENROUTER_API_KEY`,
+`--preset=openrouter`), work the same way; `kindgi providers presets` lists
+the presets (Gemini on Vertex AI too). Picking a model per
 agent, and other providers: `.claude/skills/kindgi-authoring-providers/SKILL.md`.
 
 ## The four primitives (one file each)

package/dist/templates/sample/package.json.tmpl

@@ -21,5 +21,8 @@
     "@types/node": "^22.0.0",
     "typescript": "^5.7.3",
     "vitest": "^3.0.0"
+  },
+  "allowScripts": {
+    "esbuild": false
   }
 }

package/dist/templates/sample/pnpm-workspace.yaml

@@ -3,6 +3,7 @@
 # its install script (true) or skips it (false); pnpm 11+ stops an install
 # until each one has a decision. esbuild (the CLI's bundler, and vitest's
 # transform) works without its script: its native binary comes from its
-# `@esbuild/<platform>` package, so the script stays off.
+# `@esbuild/<platform>` package, so the script stays off. npm 11+ records
+# the same decision in `package.json`'s `allowScripts`.
 allowBuilds:
   esbuild: false

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@kindgi/cli",
-  "version": "0.1.4-rc.2",
+  "version": "0.1.4-rc.4",
   "description": "Command-line interface for Kindgi™ — the sovereign AI OS. Create a pack, run it on your machine (`kindgi dev`), build, sign and deploy it, and work with a running Kindgi API from the terminal.",
   "license": "Apache-2.0",
   "repository": {
@@ -37,16 +37,16 @@
     "provenance": true
   },
   "dependencies": {
-    "@kindgi/client": "0.1.4-rc.2",
-    "@kindgi/dotenv-file": "0.1.4-rc.2",
-    "@kindgi/crypto": "0.1.4-rc.2",
-    "@kindgi/env-schema": "0.1.4-rc.2",
-    "@kindgi/flow": "0.1.4-rc.2",
-    "@kindgi/handler-runtime": "0.1.4-rc.2",
-    "@kindgi/secrets-dotenv": "0.1.4-rc.2",
-    "@kindgi/platform": "0.1.4-rc.2",
-    "@kindgi/sdk": "0.1.4-rc.2",
-    "@kindgi/types": "0.1.4-rc.2",
+    "@kindgi/client": "0.1.4-rc.4",
+    "@kindgi/dotenv-file": "0.1.4-rc.4",
+    "@kindgi/crypto": "0.1.4-rc.4",
+    "@kindgi/env-schema": "0.1.4-rc.4",
+    "@kindgi/flow": "0.1.4-rc.4",
+    "@kindgi/handler-runtime": "0.1.4-rc.4",
+    "@kindgi/secrets-dotenv": "0.1.4-rc.4",
+    "@kindgi/platform": "0.1.4-rc.4",
+    "@kindgi/sdk": "0.1.4-rc.4",
+    "@kindgi/types": "0.1.4-rc.4",
     "esbuild": "^0.24.0",
     "semver": "^7.8.5",
     "smol-toml": "^1.9.0",