docs: add AEO cross-links for agents and orchestration

[rachaelrenk]

AEO cross-link audit brief

Goal
Identify small, high-confidence internal cross-link improvements for the agents, cloud agents, and orchestration docs. Scope is limited to internal cross-links between existing pages — no new pages, rewrites, or broad SEO changes.

Source signals

• Peec snapshot (buzz/aeo-snapshots/docs/agents-orchestration/latest.json, window 2026-05-05 → 2026-06-04). High-volume prompt clusters:
  • "what are the best tools for managing multiple AI coding agents?" / "AI agent orchestration platforms" → orchestration.
  • "how do I set up AI coding agents to run on a schedule or in the background?" → scheduled agents.
  • "how do I trigger AI coding agents automatically from Slack, Linear, or GitHub?" → integrations.
• Existing-docs signal: The Cloud Agent FAQs (faqs.mdx) discuss multi-agent coordination, scheduled recurring runs, and integration-triggered triage in prose, but the page links to none of the canonical orchestration, scheduled agents, or integrations pages. This is the clearest gap in an otherwise densely cross-linked topic area.
• Google Search Console: Credentials were present in the environment, but no GSC query data was incorporated into this run; recommendations are grounded in Peec signals and existing-docs structure. The secret value was never printed, logged, or committed.

Pages touched

• src/content/docs/agent-platform/cloud-agents/faqs.mdx — added three inline cross-links in three separate Q&A sections, each pointing to a distinct canonical destination that the answer already describes.

Links added

1. FAQ "How do agents handle branching, merge conflicts, and multi-agent coordination" → Multi-agent orchestration. The answer describes "running and coordinating multiple concurrent agents" — the orchestration page is the canonical home for that model.
2. FAQ "Can cloud agents do dependency upgrades?" → Scheduled Agents. The answer says the workflow is "usually implemented as a scheduled cloud agent producing recurring runs."
3. FAQ "Can cloud agents triage issues / tickets automatically?" → Integrations. The answer references work created "via an integration trigger."

Reader next step

• A reader on the FAQ who wants to coordinate multiple agents can jump straight to the orchestration model and patterns.
• A reader interested in recurring dependency upgrades can go directly to schedule setup (CLI and web app).
• A reader exploring automated issue triage can reach the integrations setup that fires those triggers.

Open questions for human review

• Link density: all three links land on faqs.mdx. They sit in three distant Q&A sections of a hub page and point to three distinct destinations, so this stays within the audit's hub-page allowance, but reviewers may prefer to distribute some links to other pages in a follow-up.
• Two pre-existing style_lint HEADER-CASE warnings on faqs.mdx (lines 52, 81) are false positives (proper nouns: ChatGPT, Docker, Playwright) and were left untouched to keep this PR scoped to cross-linking.

Follow-up recommendations (out of scope for this pilot)

• Consider an inline orchestration link in overview.mdx's "more parallelism" bullet (currently only linked in the "Learn more" list).
• Peec UGC recommendations (Reddit/YouTube on "AI coding agents") are content-creation opportunities outside internal cross-linking.

Conversation: https://app.warp.dev/conversation/8b970481-22fc-4d40-938a-25d92b867b62
Run: https://oz.warp.dev/runs/019ebb0f-45f5-751e-b905-d3208979e5f8

This PR was generated with Oz.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
docs	Ready	Preview, Comment	Jun 12, 2026 9:07am