feat(plugin): support Copilot CLI plugin generation from collection manifests

Description

Fixes #476
Fixes #392

Added Copilot CLI plugin support, enabling HVE Core agents, prompts, instructions, and skills to be packaged as installable CLI plugins. This feature introduces YAML collection manifests that define artifact bundles, PowerShell scripts that generate plugin directories with symlinked content, a validation pipeline, and an auto-generated artifact catalog with VS Code install badges.

• feat(plugin): added 10 collection manifests (collections/*.collection.yml) defining artifact bundles for ado, coding-standards, data-science, git, github, hve-core-all, project-planning, prompt-engineering, rpi, and security-planning

• feat(plugin): implemented Generate-Plugins.ps1 to create plugin directories from collection manifests with symlinked agents, commands, instructions, and skills

  • Each plugin directory includes .github/plugin/plugin.json, a generated README.md, and symlinks to source artifacts

• feat(plugin): implemented Validate-Collections.ps1 to validate collection manifest structure, required fields, and referenced file paths against collection-manifest.schema.json

• feat(plugin): implemented Generate-Badges.ps1 to produce docs/catalog.md with browsable tables of all agents, prompts, instructions, and skills with one-click VS Code install badges

• feat(plugin): created PluginHelpers.psm1 shared PowerShell module with utility functions for path resolution, manifest parsing, CI annotation output, symlink creation, and README generation

• ci(plugin): added plugin-validation.yml reusable workflow that validates collection manifests and checks plugin directory freshness against generated output

• ci(plugin): integrated plugin validation into the existing pr-validation.yml pipeline as a required job

• feat(plugin): added plugin:generate, plugin:validate, and plugin:badges npm scripts to package.json

• docs(plugin): added CLI plugins installation guide at docs/getting-started/methods/cli-plugins.md with prerequisites, registration, and usage instructions

• docs(plugin): updated docs/getting-started/README.md to include Copilot CLI Plugins as the eighth installation method

• docs(plugin): generated docs/catalog.md artifact catalog with 21 agents, 27 prompts, 24 instructions, and 1 skill listed with install badges

• feat(plugin): generated 10 plugin directories under plugins/ with symlinked content from .github/agents/, .github/prompts/, .github/instructions/, and .github/skills/

• refactor(plugin): reordered version and description fields in plugin.json files for consistency

• style(instructions): standardized table formatting across 20+ agent, instruction, and prompt files for consistent column alignment

• feat(plugin): added collection-manifest.schema.json for validating collection YAML manifest structure

Related Issue(s)

Type of Change

Select all that apply:

Code & Documentation:

• Bug fix (non-breaking change fixing an issue)
• New feature (non-breaking change adding functionality)
• Breaking change (fix or feature causing existing functionality to change)
• Documentation update

Infrastructure & Configuration:

• GitHub Actions workflow
• Linting configuration (markdown, PowerShell, etc.)
• Security configuration
• DevContainer configuration
• Dependency update

AI Artifacts:

• Reviewed contribution with prompt-builder agent and addressed all feedback
• Copilot instructions (.github/instructions/*.instructions.md)
• Copilot prompt (.github/prompts/*.prompt.md)
• Copilot agent (.github/agents/*.agent.md)
• Copilot skill (.github/skills/*/SKILL.md)

Note for AI Artifact Contributors:

• Agents: Research, indexing/referencing other project (using standard VS Code GitHub Copilot/MCP tools), planning, and general implementation agents likely already exist. Review .github/agents/ before creating new ones.
• Skills: Must include both bash and PowerShell scripts. See Skills.
• Model Versions: Only contributions targeting the latest Anthropic and OpenAI models will be accepted. Older model versions (e.g., GPT-3.5, Claude 3) will be rejected.
• See Agents Not Accepted and Model Version Requirements.

Other:

• Script/automation (.ps1, .sh, .py)
• Other (please describe):

Sample Prompts (for AI Artifact Contributions)

AI artifact changes in this PR are limited to table formatting standardization (style-only). No new agents, prompts, or instructions were introduced.

Testing

Plugin Generation and Validation

• Ran npm run plugin:validate — all 10 collection manifests passed validation (0 errors)
• Ran npm run plugin:generate — generated 10 plugin directories with 41 agents, 54 commands, 71 instructions, and 1 skill
• Verified plugin freshness: git diff --quiet plugins/ confirmed generated output matches committed state

Copilot CLI Plugin Usage

Once installed via the plugin marketplace, artifacts become available in Copilot CLI:

Marketplace registration and plugin install:

copilot plugin marketplace add microsoft/hve-core
copilot plugin install rpi@hve-core

Agents — Available via /agent slash command or --agent flag:

# Non-interactive (one-shot)
copilot --agent=task-researcher -p "Research authentication patterns"

# Interactive (switch agents mid-session)
copilot
> /agent task-researcher
> Research authentication patterns
> /agent task-planner
> Plan based on the research

Commands — Prompts from commands/ become namespaced slash commands:

copilot
> /rpi:task-research          # runs the task-research prompt
> /git:git-commit-message     # from the git plugin bundle
> /ado:ado-get-build-info     # from the ado plugin bundle

Instructions — Included in plugin directories but not natively consumed by the CLI plugin system. Instructions apply when using the direct file approach (cloning hve-core and letting CLI auto-discover from .github/). For plugin installs, agent files reference instructions internally, so instruction guidance carries through agent behavior.

Known CLI Limitations

• ${input:variableName} template variables are VS Code-only; CLI users pass parameters as natural language
• #file: context references are VS Code-only; CLI agents use file-read tools instead
• MCP servers require separate configuration at ~/.copilot/mcp-config.json

Automated Checks

All required automated checks passed (see Required Automated Checks section below).

Checklist

Required Checks

• Documentation is updated (if applicable)
• Files follow existing naming conventions
• Changes are backwards compatible (if applicable)
• Tests added for new functionality (if applicable)

AI Artifact Contributions

• Used /prompt-analyze to review contribution
• Addressed all feedback from prompt-builder review
• Verified contribution follows common standards and type-specific requirements

Required Automated Checks

The following validation commands must pass before merging:

• Markdown linting: npm run lint:md
• Spell checking: npm run spell-check
• Frontmatter validation: npm run lint:frontmatter
• Link validation: npm run lint:md-links
• PowerShell analysis: npm run lint:ps

GHCP Artifact Maturity

All modified GHCP artifacts (agents, instructions, prompts) are existing stable files with style-only table formatting changes. No non-stable maturity levels detected.

Security Considerations

• This PR does not contain any sensitive or NDA information
• Any new dependencies have been reviewed for security issues
• Security-related scripts follow the principle of least privilege

Additional Notes

• Plugin directories under plugins/ use symlinks to source artifacts, enabling zero-copy installation and automatic content synchronization.
• The plugin-validation.yml workflow checks both manifest validity and plugin freshness (whether generated output matches current manifests).
• Collection manifests support display.ordering (manual or alpha) and display.show_badge settings for catalog generation.
• The scripts/README.md was updated to document the new scripts/plugins/ directory and its scripts.

🔧 - Generated by Copilot