agentwheel 0.14.13 → 0.16.0

package/openpack.json

@@ -1,7 +1,7 @@
 {
   "schemaVersion": 2,
   "name": "NestDevLab/agentwheel",
-  "version": "0.14.13",
+  "version": "0.16.0",
   "provides": [
     { "type": "skills", "path": "skills" }
   ]

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "agentwheel",
-  "version": "0.14.13",
+  "version": "0.16.0",
   "description": "Weave skills, rules, and instructions across every AI agent.",
   "type": "module",
   "license": "MIT",

package/skills/agentwheel/SKILL.md

@@ -1,16 +1,16 @@
 ---
 name: agentwheel
-description: Use agentwheel to discover, add, install, update, customize, eject, and uninstall agent skills, rules, instructions, commands, MCP, hooks, settings, and plugin artifacts across runtimes.
+description: Discover or manage reusable agent artifacts with Agentwheel. Use when a requested agent capability, integration, workflow, policy, or tool may already exist, and when adding, installing, updating, customizing, ejecting, or uninstalling skills, rules, instructions, commands, MCP, hooks, settings, plugins, or subagents across runtimes.
 allowed-tools: [Bash]
 license: MIT
 metadata:
   author: NestDevLab
-  version: "0.14.13"
+  version: "0.16.0"
 ---
 
 # agentwheel
 
-Use this skill when a user wants to install, manage, update, remove, or inspect agent skills or other agentwheel-managed artifacts.
+Agentwheel discovers reusable artifacts and manages their desired state across runtimes.
 
 agentwheel is the control plane. It reads packages from sources, stores desired state in `.agentwheel/config.json`, plans runtime changes, and writes only through `install`. Treat runtime output directories as generated files.
 
@@ -29,11 +29,12 @@ Mental model:
 - Gmail, Drive, registry publishing, git commits, pushes, and runtime reloads/restarts are separate external side effects. Get explicit approval for them.
 - Programmatic adapters execute local code. Use `--adapter-module` only with `--allow-adapter-code` after the user approves that local code execution.
 - OpenClaw plugin artifacts are only planned by default. Use `--execute-plugins` only after explicit approval.
+- Search results are proposals, not approval. Never add, install, enable, or change configuration until the user confirms the artifact and target scope.
 
 ## Core Flow
 
 ```bash
-agentwheel registry search tmux
+agentwheel search tmux
 agentwheel add github:NestDevLab/agent-mesh --skill codex-tmux --adapter codex --installation-type local --mode tracking
 agentwheel plan
 agentwheel install
@@ -72,14 +73,34 @@ agentwheel init package
 
 `agentwheel init package` creates `openpack.json`, `instructions/`, `rules/`, `skills/`, and `instructions/AGENTS.md`.
 
-## Discovery
+## Discovery And Recommendations
 
-Refresh and search the optional registry:
+Search configured registries and public catalogue sources:
+
+```bash
+agentwheel search "<query>"
+agentwheel search "<query>" --json --limit 10
+```
+
+When a reusable artifact could satisfy the request:
+
+1. Extract the capability and constraints from the complete request.
+2. Generate one to four short lexical queries using capability terms, synonyms, runtime names, and artifact types. Prefer English catalogue terms when the request uses another language.
+3. Run one `agentwheel search "<query>" --json --limit 10` per variant. Stop after four calls; do not recursively refine without new user requirements.
+4. Merge results by stable `id`. Treat CLI scores as retrieval signals, not semantic confidence.
+5. Rerank against the original request using capabilities, runtime or ecosystem, artifact type, description, tags, `provides`, and installability. Do not infer capabilities absent from result metadata.
+6. Suggest zero to three distinct artifacts. For each, give its name or source, one evidence-based match reason, installability, and a safe next command.
+7. Wait for explicit approval before `add`, `install`, plugin execution, or configuration changes.
+
+For automatic suggestions, skip discovery when the user explicitly wants custom implementation, has already selected an artifact, an installed artifact clearly satisfies the request, candidates are only weak lexical matches, or the same suggestion was already shown without new evidence.
+
+Search recommendations are conversational only: they do not select OpenPack `suggests`, mutate desired state, or imply installation approval.
+
+Registry maintenance remains explicit:
 
 ```bash
 agentwheel registry update
 agentwheel registry list
-agentwheel registry search <query>
 ```
 
 Inspect an explicit source before adding it:
@@ -91,7 +112,7 @@ agentwheel list ./local-agent-pack
 agentwheel scan ./local-agent-pack
 ```
 
-Filter discovery to specific artifacts:
+Filter source inspection to specific artifacts:
 
 ```bash
 agentwheel list github:owner/repo --select skills/review --select rules/core.md
@@ -498,11 +519,11 @@ If a selected artifact is missing:
 agentwheel list <source>
 ```
 
-If registry short names fail:
+If a registry short name fails:
 
 ```bash
 agentwheel registry update
-agentwheel registry search <query>
+agentwheel search "<query>" --scope registry
 ```
 
 If npm update checks are noisy: