Add AI discovery endpoints to docs site

[Copilot]

GEO Audit scored 0/6 on AI discovery endpoints. Adds four static files to docs/public/ (served as-is by Astro) to address this.

New files

• docs/public/.well-known/ai.txt — AI crawler permissions; points to llms.txt and ai/summary.json
• docs/public/ai/summary.json — project name, description, docs URL, source repo, install command
• docs/public/ai/faq.json — 10 Q&A entries covering installation, engines, safe outputs, MCP, debugging, cache-memory
• docs/public/ai/service.json — capabilities list, supported engines, GitHub integrations, license

Infra

• Added !docs/public/.well-known/ai.txt exception to .gitignore (root *.txt rule was swallowing it, same pattern as robots.txt)

[github-actions]

Hey @copilot-swe-agent 👋 — nice work addressing the GEO Audit gap by adding the AI discovery endpoints! The four static files (ai.txt, summary.json, faq.json, service.json) are well-structured and the .gitignore exception follows the existing robots.txt pattern perfectly.

One small gap before this is ready to merge:

• No tests — the diff doesn't include any test coverage. For static content files like these, unit tests may not be meaningful, but if there's a docs build/deploy test suite or a CI check that validates the presence and schema of these well-known files, it would be worth hooking into that. If no such suite exists, a brief note in the PR body explaining why tests aren't applicable here would help reviewers move faster.

If you'd like a hand, you can assign this prompt to your coding agent:

Check whether the docs site (docs/) has any existing test or validation scripts (e.g. in package.json scripts, a Makefile, or a CI workflow) that verify static public files or JSON schema.
If a validation mechanism exists, add checks for the new files:
  - docs/public/.well-known/ai.txt (must be non-empty, must contain "Allow:" and "Llms-txt:" lines)
  - docs/public/ai/summary.json (must be valid JSON with keys: name, description, docs, source, install)
  - docs/public/ai/faq.json (must be valid JSON with a non-empty "faq" array)
  - docs/public/ai/service.json (must be valid JSON with keys: service, capabilities, engines, install)
If no validation mechanism exists, add a comment to the PR body explaining why tests are not applicable for these static content files.

Generated by Contribution Check · ● 6.7M · ◷

[Copilot]

Pull request overview

This PR adds machine-readable “AI discovery” endpoints to the docs site (served from docs/public/) and updates ignore rules so the new ai.txt file is committed.

Changes:

• Add ai/summary.json, ai/service.json, and ai/faq.json with project metadata, capabilities, and Q&A.
• Add .well-known/ai.txt to advertise crawler permissions and link to the summary/llms resources.
• Update root .gitignore to unignore docs/public/.well-known/ai.txt despite a global *.txt ignore.

Show a summary per file

File	Description
docs/public/ai/summary.json	Adds project summary metadata for AI consumers.
docs/public/ai/service.json	Adds machine-readable capabilities/engines/integrations metadata.
docs/public/ai/faq.json	Adds a structured FAQ for AI consumers.
docs/public/.well-known/ai.txt	Adds AI crawler permissions + pointers to discovery resources.
.gitignore	Unignores the new ai.txt file under docs/public/.well-known/.

Copilot's findings

Tip

Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

• Files reviewed: 4/5 changed files
• Comments generated: 3