feat: upgraded to the v2 schema for sbx mixins

Adds the identity manifest and rewrites the README around the current
install surface: --install with owner/repo shorthand and --git-host,
provenance-backed --list-bundles / --update-bundle / --uninstall, and
the update semantics for files and mcp entries.

README.md

@@ -1,56 +1,116 @@
-# coyote-config-template
+# coyote-bundle-template
 
-A starter template for sharing [Coyote](https://github.com/Dark-Alex-17/coyote)
+A starter template for [Coyote](https://github.com/Dark-Alex-17/coyote)
-configurations via any Git repository. Repositories structured like this enable
+bundles: shareable configurations distributed via any Git repository.
-users to share agents, roles, macros, tools, and MCP servers in Coyote easily.
+Repositories structured like this let users share agents, roles, skills,
+macros, tools, and MCP servers in Coyote easily. Bundles are Coyote's
+equivalent of plugins.
 
 Fork this repo, customize the assets to your taste, then install your fork
 into Coyote with a single command.
 
 ## Quick start
 
-Install everything in this template into your local Coyote config:
+Install everything in this template into your local Coyote config using the
+`owner/repo` shorthand (expanded against github.com by default):
 
 ```sh
-coyote --install-from https://github.com/<you>/coyote-config-template
+coyote --install <you>/coyote-bundle-template
 ```
 
-or from within the Coyote REPL:
+or with a full Git URL, or from within the Coyote REPL:
+
+```sh
+coyote --install https://github.com/<you>/coyote-bundle-template
+```
 
 ```
-.install remote https://github.com/<you>/coyote-config-template
+.install <you>/coyote-bundle-template
+```
+
+Hosting somewhere other than GitHub? Point the shorthand at your forge:
+
+```sh
+coyote --install <group>/<subgroup>/coyote-bundle-template --git-host gitlab.com
 ```
 
 Pin to a specific branch, tag, or commit by suffixing `#<ref>`:
 
 ```sh
-coyote --install-from https://github.com/<you>/coyote-config-template#v1.0.0
+coyote --install <you>/coyote-bundle-template#v1.0.0
-coyote --install-from https://github.com/<you>/coyote-config-template#main
+coyote --install <you>/coyote-bundle-template#main
-coyote --install-from https://github.com/<you>/coyote-config-template#abc1234
+coyote --install <you>/coyote-bundle-template#abc1234
 ```
 
 Restrict the install to a single asset category with `--filter`:
 
 ```sh
-coyote --install-from https://github.com/<you>/coyote-config-template --filter agents
+coyote --install <you>/coyote-bundle-template --filter agents
-coyote --install-from https://github.com/<you>/coyote-config-template --filter mcp_config
+coyote --install <you>/coyote-bundle-template --filter mcp-config
 ```
 
-Valid filter values: `agents`, `roles`, `skills`, `macros`, `functions`, `mcp_config`.
+Valid filter values: `agents`, `roles`, `skills`, `macros`, `functions`, `mcp-config`.
 
 Skip per-file conflict prompts with `--install-force`:
 
 ```sh
-coyote --install-from https://github.com/<you>/coyote-config-template --install-force
+coyote --install <you>/coyote-bundle-template --install-force
 ```
 
+## Managing the installed bundle
+
+Coyote records everything a bundle install writes (file paths, MCP server
+entries, and content hashes) in a provenance store, which powers the full
+lifecycle:
+
+```sh
+coyote --list-bundles                          # name, version, source, drift
+coyote --install <bundle-name>                 # update by installed name
+coyote --update-bundle <bundle-name>           # update from the recorded source
+coyote --update-bundle <bundle-name>#<ref>     # move a pin while updating
+coyote --update-bundle <bundle-name> --yes     # non-interactive: keep local edits, refresh the rest
+coyote --uninstall <bundle-name>               # remove owned files and MCP entries
+coyote --uninstall <bundle-name> --yes         # skip the confirmation prompt
+```
+
+The same operations are available in the REPL as `.list bundles`,
+`.install <bundle-name>`, and `.uninstall <bundle-name> [--yes]`.
+
+Updates and uninstalls never clobber your local edits: files you modified
+after install are prompted for (and kept by default on uninstall), while
+untouched files refresh or delete silently. See the
+[Sharing Configurations wiki](https://github.com/Dark-Alex-17/coyote/wiki/Sharing-Configurations)
+for the full ownership and drift model.
+
+## The manifest (`coyote-bundle.yaml`)
+
+The optional manifest at the repo root declares the bundle's identity, and
+nothing else. Coyote discovers installable content by scanning the asset
+directories; the manifest never lists files.
+
+```yaml
+name: coyote-bundle-template
+version: "1.0.0"
+description: Starter template for Coyote bundles
+homepage: https://github.com/Dark-Alex-17/coyote-bundle-template
+```
+
+- `name` is how consumers refer to the bundle in `--update-bundle` and
+  `--uninstall`. Ship a manifest so your bundle keeps a stable name across
+  forks; without one, Coyote derives the name from the repository name.
+- `version` shows up in `--list-bundles` (falls back to the short commit SHA).
+- If your chosen name collides with another installed bundle or a built-in
+  asset category, Coyote qualifies it with the repo owner (`owner/name`)
+  instead of overwriting anything.
+
 ## Layout
 
-Coyote only reads these top-level directories. Anything else in the repo is
+Coyote only reads the manifest and these top-level directories. Anything
-ignored.
+else in the repo is ignored.
 
 ```
-coyote-config-template/
+coyote-bundle-template/
+├── coyote-bundle.yaml             # Optional identity manifest
 ├── agents/
 │   └── <agent-name>/
 │       ├── config.yaml            # LLM-loop agent
@@ -65,26 +125,28 @@ coyote-config-template/
 │       └── SKILL.md               # Skill with frontmatter + body
 ├── macros/
 │   └── <macro-name>.yaml          # Positional/rest-args + REPL command steps
+├── mcp.json                       # MCP server config (merged with local)
 └── functions/
     ├── tools/
     │   └── *.sh / *.py / *.ts     # Global tools (auto chmod +x on install)
-    └── mcp.json                   # MCP server config (merged with local)
 ```
 
-The `functions/mcp.json` file is **merged** into your existing local file
+The `mcp.json` file is **merged** into your existing local file
 on install (not overwritten). For conflicting server names, you'll be
 prompted to keep yours, take the remote's, or rename the remote entry.
+On update, entries the bundle installed and you never edited take the
+remote side automatically.
 
 ## What's in this template
 
-| Asset  | File                              | What it is                                              |
+| Asset | File                             | What it is                                              |
-|--------|-----------------------------------|---------------------------------------------------------|
+|-------|----------------------------------|---------------------------------------------------------|
-| Agent  | `agents/hello-agent/config.yaml`  | Tiny LLM-loop agent that greets the user.               |
+| Agent | `agents/hello-agent/config.yaml` | Tiny LLM-loop agent that greets the user.               |
-| Role   | `roles/explainer.md`              | Role that explains technical concepts simply.           |
+| Role  | `roles/explainer.md`             | Role that explains technical concepts simply.           |
-| Skill  | `skills/rust-fmt/SKILL.md`        | Skill demonstrating `enabled_tools` + `auto_unload`.    |
+| Skill | `skills/rust-fmt/SKILL.md`       | Skill demonstrating `enabled_tools` + `auto_unload`.    |
-| Macro  | `macros/greet.yaml`               | Macro showing positional and rest-arg variables.        |
+| Macro | `macros/greet.yaml`              | Macro showing positional and rest-arg variables.        |
-| Tool   | `functions/tools/greet.sh`        | Bash tool using Coyote's argc-style annotations.          |
+| Tool  | `functions/tools/greet.sh`       | Bash tool using Coyote's argc-style annotations.        |
-| MCP    | `functions/mcp.json`              | One vanilla server + one with a vault secret reference. |
+| MCP   | `mcp.json`                       | One vanilla server + one with a vault secret reference. |
 
 Each sample is intentionally minimal. Replace it with your own work, or
 delete what you don't need.
@@ -122,16 +184,41 @@ Bash, Python, and TypeScript scripts are auto-detected and given the
 executable bit on install.
 
 ### MCP servers
-Add or modify entries in `functions/mcp.json`
+Add or modify entries in `mcp.json`
 ([MCP Servers wiki](https://github.com/Dark-Alex-17/coyote/wiki/MCP-Servers)).
 Use `{{SECRET_NAME}}` placeholders for values you don't want to commit;
 Coyote will detect missing secrets after the merge and prompt you to add
 them to the vault (or list them for you to add via `coyote --add-secret`).
 
+### (Optional) Sandbox mixins (`sbx-mixin.yaml`)
+If consumers of your bundle run Coyote in [Sandbox mode](https://github.com/Dark-Alex-17/coyote/wiki/Sandboxes),
+they'll need any external binaries and network domain allowances declared
+in an `sbx-mixin.yaml` file. Coyote auto-discovers mixin files at known
+locations on every `coyote --sandbox` invocation, no flags required.
+
+This template ships two starter examples:
+
+- **`agents/hello-agent/sbx-mixin.yaml`:** Per-agent mixin, applied when
+  the agent is installed and any `coyote --sandbox` runs.
+- **`functions/sbx-mixin.yaml`:** Shared by all custom tools in this
+  bundle. For per-tool granularity, use `functions/<tool>/sbx-mixin.yaml`
+  instead.
+
+Both starters are commented-out templates. Simply open them and fill in the
+domains and install commands your assets actually need. Delete the files
+if they're not needed.
+
+> ⚠️ **Privilege grant.** Anyone installing your bundle is granting the
+> mixins' install commands (passwordless sudo) and network domain
+> allowances inside their sandboxes. Document any non-obvious entries in
+> this README so they don't have to grep your YAML to find out what
+> they're accepting. See the [Sharing Configurations - Sandbox Implications](https://github.com/Dark-Alex-17/coyote/wiki/Sharing-Configurations#sandbox-implications)
+> wiki page for the full security model.
+
 ## Secrets workflow
 
 Anywhere you reference a secret in `mcp.json` (or in any installed file),
-use the `{{NAME}}` placeholder syntax. After `--install-from` completes:
+use the `{{NAME}}` placeholder syntax. After an install completes:
 
 - **Interactive mode**: Coyote prompts you per-secret to add the value to
   the vault. On the first "Yes," it creates the vault password file if
@@ -145,9 +232,15 @@ the full secrets workflow.
 
 ## Tips for forks
 
+- Ship a `coyote-bundle.yaml` with your own `name` so your bundle keeps a
+  stable identity no matter what consumers or forks call the repository.
 - Pin your fork to tagged releases so consumers can install with
-  `#<tag>` for reproducibility.
+  `#<tag>` for reproducibility, and bump the manifest `version` alongside
+  your tags so `coyote --list-bundles` shows something meaningful.
 - Keep agent-local logic in `agents/<name>/scripts/`. Global
   tools (in `functions/tools/`) are shared across every agent.
-- The `mcp.json` merge is *additive*. If you remove a server from your
+- Removing a file from your fork does propagate: on the next
-  fork, existing installs of that server are _not_ pruned. That's by design.
+  `--update-bundle`, consumers are asked whether to keep or delete files
+  the bundle no longer ships. Removing an `mcp.json` server does not prune
+  existing installs of that entry; uninstalling the bundle removes the
+  entries it added (unless the user edited them).

agents/hello-agent/sbx-mixin.yaml

@@ -0,0 +1,24 @@
+schemaVersion: '2'
+kind: mixin
+name: agent-hello
+description: >
+  Example sbx mixin for the hello-agent. Coyote auto-discovers and applies
+  any sbx-mixin.yaml co-located with an agent whenever you run
+  `coyote --sandbox`. Use this to declare any binaries the agent needs to
+  install inside the sandbox and any network domains it needs to reach.
+  Delete this file if your agent doesn't need extra sandbox setup.
+
+permissions:
+  network:
+    allowedDomains:
+      # Replace with the domains your agent actually hits.
+      # Example: an agent that calls api.example.com would add:
+      # - "api.example.com:443"
+
+setup:
+  install:
+    # Replace with any binaries your agent depends on. Runs as the
+    # `agent` user (UID 1000) with passwordless sudo. Example:
+    # - command: "sudo apt-get update && sudo apt-get install -y httpie"
+    #   user: "1000"
+    #   description: Install httpie for hello-agent's API calls

coyote-bundle.yaml

@@ -0,0 +1,7 @@
+# Bundle identity (optional). Without this file, Coyote derives the bundle
+# name from the repository name. The manifest never lists files: everything
+# Coyote installs is discovered by scanning the asset directories.
+name: coyote-bundle-template
+version: "1.0.0"
+description: Starter template for Coyote bundles
+homepage: https://github.com/Dark-Alex-17/coyote-bundle-template

functions/sbx-mixin.yaml

@@ -0,0 +1,26 @@
+schemaVersion: '2'
+kind: mixin
+name: shared-custom-tools
+description: >
+  Example global sbx mixin that applies to every Coyote sandbox you launch.
+  Use this for binaries and network domains shared by multiple custom tools
+  in functions/tools/. For per-tool needs that aren't shared, create
+  functions/<tool-name>/sbx-mixin.yaml instead (per-tool mixins are also
+  auto-discovered).
+
+  Delete this file if your fork doesn't need extra sandbox setup.
+
+permissions:
+  network:
+    allow:
+      # Replace with the domains your custom tools reach.
+      # Example: a tool that calls a private API:
+      # - "api.your-company.com:443"
+
+setup:
+  install:
+    # Replace with binaries your custom tools depend on. Runs as the
+    # `agent` user (UID 1000) with passwordless sudo. Example:
+    # - command: "sudo apt-get update && sudo apt-get install -y httpie xmlstarlet"
+    #   user: "1000"
+    #   description: Install custom tool dependencies