richfrem/agent-plugins-skills / plugins/agent-extension-authoring/.apm/skills/packaging-copilot-cowork-plugins
packaging-copilot-cowork-plugins
VerifiedCreates, validates and prepares Microsoft 365 Copilot Cowork plugin App Packages containing Agent Skills and optional remote MCP connectors. Use when scaffolding, converting, validating, provisioning, packaging or sharing a Cowork plugin. Not for APM package creation or installation (use packaging-with-apm).
Packaging Copilot Cowork Plugins
Creates and validates Microsoft 365 app packages for Copilot Cowork plugins. A package contains the plugin manifest, icons, and one or more skills; it may also declare remote MCP connectors.
Contents
Constraints
- Package the plugin as an M365 App Package
.zip; do not treat each skill as a separate app package. - Keep each skill portable and self-contained. In tests, Cowork and Microsoft 365 Copilot Chat ran a packaged Python script and returned its output file; Microsoft does not guarantee this. Prefer scripts that write files (a local server is unreachable) and put must-run functionality behind a remote MCP connector.
- Publish rejects some file types (e.g.
.mmd) and allows 20 companion files per skill; use the scripts. - WIQD is preview software. Check its current official documentation and exact installed version, then pin the tested version in repository or CI setup. Do not guess a version or assume command compatibility.
- Do not authenticate, provision, share, publish, or delete tenant resources without explicit user approval for the target tenant and scope. Never request or store passwords, tokens, or client secrets in skill files or package artifacts.
- Inspect existing files before editing. Do not overwrite source skills, manifests, icons, or environment configuration without approval.
Quick start
Identify whether the source is a WIQD project or a Claude/Open Plugin project, and read the package lifecycle for the route. Without WIQD, build locally from an APM plugin in two steps:
python scripts/scaffold_cowork_package.py --apm-plugin <plugin> --output <dir>/appPackage \
--developer-name <name> --website-url <url> --privacy-url <url> --terms-url <url>
python scripts/package_cowork_plugin.py --plugin-path <dir>/appPackage --output-path <output.zip>
The scaffold bundles over-limit skills and filters unsupported file types; the packager
checks the manifest, SKILL.md frontmatter, links and file types.
Workflow
- Reuse supplied requirements. Confirm the plugin name, skills to include, connector endpoint and auth expectations, privacy/terms URLs for conversion when needed, and whether the user wants only a local ZIP or tenant-side actions.
- Follow the relevant lifecycle route in the package lifecycle. When building from scratch, follow its source-authoring route before creating the M365 package. Keep generated package files under the selected project; do not copy server executables into the app package.
- Validate the built ZIP and inspect its contents before delivery. Report the exact artifact path, validation result, tool version, and any unperformed tenant actions.
- If publish validation rejects something the checks missed, fix the tools here and rebuild.
Verification
Check that the ZIP has a valid manifest, required icons, and exactly the intended skill folders and connector declarations; ensure it contains no credentials, local secrets, or unintended files. Verify deep validation applies to the final artifact, not only an earlier draft. Do not report tenant sharing or publication as complete unless it was explicitly approved and confirmed. See acceptance criteria for the completion contract.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 469k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 321k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 321k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 321k | Updated | Claude CodeCodex | Verified | Install |