feat: add six public skill packets

marketplaces/default.json

@@ -485,6 +485,84 @@
                 "sample"
             ]
         },
+        {
+            "name": "openui-workspace-delivery",
+            "source": "./openui-workspace-delivery",
+            "description": "Teach an agent how to install OpenUI MCP Studio, connect it to a host, and run the core UI generation and review workflow through a local MCP lane.",
+            "category": "productivity",
+            "keywords": [
+                "openui",
+                "mcp",
+                "ui-generation",
+                "local-first",
+                "review"
+            ]
+        },
+        {
+            "name": "movi-review-first-bundle",
+            "source": "./movi-review-first-bundle",
+            "description": "Teach an agent how to install Movi, stay review-first, and use the safest manifest and batch-analysis tools before deeper mutation.",
+            "category": "productivity",
+            "keywords": [
+                "movi",
+                "mcp",
+                "review-first",
+                "media",
+                "batch"
+            ]
+        },
+        {
+            "name": "switchyard-runtime-diagnostics",
+            "source": "./switchyard-runtime-diagnostics",
+            "description": "Teach an agent how to connect Switchyard and use the safest runtime diagnostics and provider-boundary tools first.",
+            "category": "integration",
+            "keywords": [
+                "switchyard",
+                "mcp",
+                "diagnostics",
+                "runtime",
+                "read-only"
+            ]
+        },
+        {
+            "name": "prooftrail-mcp",
+            "source": "./prooftrail-mcp",
+            "description": "Teach an agent how to install ProofTrail as a governed stdio MCP surface and start with read-first proof and recovery workflows.",
+            "category": "productivity",
+            "keywords": [
+                "prooftrail",
+                "mcp",
+                "browser-evidence",
+                "recovery",
+                "local-first"
+            ]
+        },
+        {
+            "name": "shopflow-read-only-packet",
+            "source": "./shopflow-read-only-packet",
+            "description": "Teach an agent how to connect Shopflow for read-only submission-readiness and storefront packet review without live store claims.",
+            "category": "productivity",
+            "keywords": [
+                "shopflow",
+                "mcp",
+                "read-only",
+                "distribution",
+                "packet"
+            ]
+        },
+        {
+            "name": "prooflane-mcp",
+            "source": "./prooflane-mcp",
+            "description": "Teach an agent how to install Prooflane, connect the repo-native MCP server, and inspect governed automation and proof workflows safely.",
+            "category": "productivity",
+            "keywords": [
+                "prooflane",
+                "mcp",
+                "automation",
+                "proof",
+                "review"
+            ]
+        },
         {
             "name": "automation-creation",
             "source": "./plugins/automation-creation",

skills/movi-review-first-bundle/README.md

@@ -0,0 +1,69 @@
+# Movi Review-First Bundle Public Skill
+
+Status: `clawhub-listed-live`
+
+This folder is the public, self-contained skill packet for Movi.
+It is meant to travel into OpenHands- or ClawHub-style review flows without
+forcing the reviewer to reopen the whole repo first.
+
+## Purpose
+
+Use it when you want one portable skill folder that teaches five things inside
+the packet itself:
+
+- what Movi helps an agent do
+- how to install and attach the local Movi MCP server
+- which tools are safe first for review-first batch work
+- what one good first-success path looks like
+- what the packet must not claim
+
+## What this packet includes
+
+- `SKILL.md`
+  - the agent-facing workflow brief
+- `manifest.yaml`
+  - packet metadata and truthful listing boundary
+- `references/README.md`
+  - quick map of the local reference files
+- `references/INSTALL.md`
+  - install and attach walkthrough
+- `references/OPENHANDS_MCP_CONFIG.json`
+  - host config snippet for OpenHands-style `mcpServers`
+- `references/OPENCLAW_MCP_CONFIG.json`
+  - host config snippet for OpenClaw-style `mcp.servers`
+- `references/CAPABILITIES.md`
+  - exact Movi tool surface and safe-first order
+- `references/DEMO.md`
+  - first-success prompt plus expected tool sequence
+- `references/TROUBLESHOOTING.md`
+  - attach, proof, and runtime failure checks
+
+## Best-fit hosts
+
+- OpenHands/extensions contribution flow
+- ClawHub-style skill publication
+- repo-local skill import flows that expect one standalone folder with its own
+  install, capability, and demo notes
+
+## Current verified state
+
+- the ClawHub skill listing is live
+- the OpenHands/extensions lane is still folder-ready, not merged
+
+## What this packet must not claim
+
+- no live OpenHands listing without fresh PR/read-back
+- no browser-extension marketplace listing
+- no hosted Movi SaaS or hidden execute shortcut
+
+## Existing repo-owned helper files
+
+This packet still ships the repo-owned helper files below because they are
+useful during review:
+
+- `codex.mcp.json`
+- `claude-code.mcp.json`
+- `openclaw.mcp.json`
+- `install-and-proof.md`
+
+They are helper artifacts, not evidence that a live listing already exists.

skills/movi-review-first-bundle/SKILL.md

@@ -0,0 +1,79 @@
+---
+name: movi-review-first-bundle
+description: Teach an agent to install Movi's local MCP server, stay review-first, and use the safest manifest and batch-analysis tools before deeper mutation.
+version: 1.0.0
+triggers:
+  - movi
+  - movi organizer
+  - review-first batch
+  - manifest review
+  - openclaw movi
+---
+
+# Movi Review-First Bundle
+
+Teach the agent how to install, connect, and use Movi as a local-first
+review-first MCP workflow.
+
+## Use this skill when
+
+- the user wants to inspect one batch or review queue before executing anything
+- the host can run a local MCP server from a repo checkout
+- the operator wants a truthful packet that explains install, attach, proof, and
+  safe-first usage without claiming a live listing
+
+## What this package teaches
+
+- how to wire Movi MCP into Codex, Claude Code, OpenHands, or OpenClaw
+- which Movi tools are safe first when the work must stay review-first
+- how to inspect jobs, manifests, and review rules before calling heavier
+  mutation tools
+- how to keep listing claims honest while still proving the packet is real
+
+## Start here
+
+1. Read [references/INSTALL.md](references/INSTALL.md)
+2. Load the right host config from:
+   - [references/OPENHANDS_MCP_CONFIG.json](references/OPENHANDS_MCP_CONFIG.json)
+   - [references/OPENCLAW_MCP_CONFIG.json](references/OPENCLAW_MCP_CONFIG.json)
+3. Skim the tool surface in [references/CAPABILITIES.md](references/CAPABILITIES.md)
+4. Run the first-success path in [references/DEMO.md](references/DEMO.md)
+
+## Safe-first workflow
+
+1. `jobs.list`
+2. `review_queue.get`
+3. `manifest.get`
+4. `analyze.create`
+5. only then consider preview or patch-style actions such as:
+   - `manifest.patch_row`
+   - `manifest.batch_patch`
+   - `review_rule.preview`
+
+## Suggested first prompt
+
+Use Movi to inspect the current review-first workload. Start with `jobs.list`,
+`review_queue.get`, and `manifest.get`. Summarize which batch needs attention
+first. If the manifest looks stable, use `analyze.create` to produce one
+analysis artifact. Do not call `manifest.patch_row`, `manifest.batch_patch`, or
+`review_rule.apply` unless I explicitly ask for a patch or rule change.
+
+## Success checks
+
+- the host can launch the local Movi MCP server from the provided config
+- the packet proves one real job/review queue exists instead of describing an
+  imaginary batch
+- the first analysis artifact is tied to a real manifest or job record
+
+## Boundaries
+
+- Movi stays a local-first review-first MCP workflow, not a hosted SaaS
+- this packet does not claim a live OpenHands or ClawHub listing
+- this packet does not bypass `review-first -> dry-run -> execute`
+
+## Local references
+
+- [references/INSTALL.md](references/INSTALL.md)
+- [references/CAPABILITIES.md](references/CAPABILITIES.md)
+- [references/DEMO.md](references/DEMO.md)
+- [references/TROUBLESHOOTING.md](references/TROUBLESHOOTING.md)

skills/movi-review-first-bundle/claude-code.mcp.json

@@ -0,0 +1,10 @@
+{
+  "mcpServers": {
+    "movi": {
+      "command": "bash",
+      "args": [
+        "/absolute/path/to/movi-organizer/tooling/runtime/run_mcp_stdio.sh"
+      ]
+    }
+  }
+}

skills/movi-review-first-bundle/codex.mcp.json

@@ -0,0 +1,10 @@
+{
+  "mcpServers": {
+    "movi": {
+      "command": "bash",
+      "args": [
+        "/absolute/path/to/movi-organizer/tooling/runtime/run_mcp_stdio.sh"
+      ]
+    }
+  }
+}

skills/movi-review-first-bundle/install-and-proof.md

@@ -0,0 +1,20 @@
+# Movi Agent Bundle Install And Proof
+
+Status: `submission-ready-unlisted`
+
+## Install
+
+- Codex: use `examples/skills/codex.mcp.json`
+- Claude Code: use `examples/skills/claude-code.mcp.json`
+- OpenClaw-style hosts: use `examples/skills/openclaw.mcp.json`
+
+## Proof Loop
+
+1. `bash tooling/runtime/bootstrap_env.sh`
+2. `npm run mcp:tools`
+3. `npm run public:readiness`
+4. `bash tooling/gates/verify_repo_final.sh`
+
+## Boundary
+
+This proves the repo-owned bundle exists. It does not prove a live listing.

skills/movi-review-first-bundle/manifest.yaml

@@ -0,0 +1,36 @@
+version: 1
+id: movi-agent-bundle
+display_name: Movi Review-First Bundle
+status: clawhub-listed-live
+canonical_repo: xiaojiou176-open/movi-organizer
+entrypoint: SKILL.md
+package_shape: skill-folder
+registry_targets:
+  clawhub:
+    status: listed-live
+    package_shape: skill-folder
+    listing_url: https://www.clawhub.ai/skills/movi-review-first-bundle
+    submit_via: clawhub publish . --slug movi-review-first-bundle --name "Movi Review-First Bundle" --version 1.0.0 --tags mcp,review-first,batch,media,movi
+  openhands-extensions:
+    status: folder-ready
+    package_shape: skill-folder
+    submit_via: submit this folder as skills/movi-review-first-bundle/ in OpenHands/extensions
+bundle_surfaces:
+  - SKILL.md
+  - README.md
+  - references/README.md
+  - references/INSTALL.md
+  - references/CAPABILITIES.md
+  - references/DEMO.md
+  - references/TROUBLESHOOTING.md
+  - references/OPENHANDS_MCP_CONFIG.json
+  - references/OPENCLAW_MCP_CONFIG.json
+  - codex.mcp.json
+  - claude-code.mcp.json
+  - openclaw.mcp.json
+  - install-and-proof.md
+claims_not_made:
+  - no live OpenHands listing is claimed
+  - no live OpenClaw listing is claimed
+  - no browser-extension marketplace listing is claimed
+  - no hosted Movi SaaS claim is made

skills/movi-review-first-bundle/openclaw.mcp.json

@@ -0,0 +1,10 @@
+{
+  "mcpServers": {
+    "movi": {
+      "command": "bash",
+      "args": [
+        "/absolute/path/to/movi-organizer/tooling/runtime/run_mcp_stdio.sh"
+      ]
+    }
+  }
+}

skills/movi-review-first-bundle/references/CAPABILITIES.md

@@ -0,0 +1,36 @@
+# Movi MCP Capabilities
+
+These are the current repo-owned Movi MCP tools exposed by `npm run mcp:tools`.
+
+## Best first tools
+
+1. `jobs.list`
+   - inspect the current batch queue before choosing a target
+2. `review_queue.get`
+   - see what is waiting for review-first attention
+3. `manifest.get`
+   - inspect the current manifest state before proposing any patch
+4. `analyze.create`
+   - create one analysis artifact without mutating the batch
+
+## Useful follow-through tools
+
+- `jobs.get`
+- `report.get`
+- `strategy_packs.list`
+- `watch_sources.list`
+- `inbox.scan`
+- `inbox.analyze`
+
+Use these after the first-safe picture is already clear.
+
+## Heavier mutation tools
+
+- `manifest.patch_row`
+- `manifest.batch_patch`
+- `review_rule.preview`
+- `review_rule.apply`
+- `apply.preview`
+
+Treat these as gated actions. Do not run them in the first pass unless the user
+explicitly asks for a patch or rule change.

skills/movi-review-first-bundle/references/DEMO.md

@@ -0,0 +1,33 @@
+# First-Success Path
+
+This is the shortest demo that proves the skill does real work instead of only
+describing a workflow.
+
+## Demo prompt
+
+Use Movi to inspect the current review-first batch queue. Start with
+`jobs.list`, `review_queue.get`, and `manifest.get`. Summarize which batch needs
+attention first. If the manifest is present and stable, run `analyze.create` to
+produce one analysis artifact. Stop before `manifest.patch_row`,
+`manifest.batch_patch`, or `review_rule.apply` unless I explicitly ask for a
+change.
+
+## Expected tool sequence
+
+1. `jobs.list`
+2. `review_queue.get`
+3. `manifest.get`
+4. `analyze.create`
+
+## Visible success criteria
+
+- the host attaches the local Movi MCP server
+- the agent names at least one real job or review queue item
+- the analysis step points back to a real manifest or batch artifact
+- the agent keeps the workflow review-first instead of jumping into mutation
+
+## What to check if it fails
+
+1. the host config path still points at the real repo checkout
+2. `bash tooling/runtime/bootstrap_env.sh` was run
+3. `npm run mcp:tools` returns the tool inventory without errors