Skip to content

docs: add AEO cross-links for agents and orchestration#183

Closed
rachaelrenk wants to merge 2 commits into
mainfrom
docs/aeo-crosslinks-agents-orchestration-2
Closed

docs: add AEO cross-links for agents and orchestration#183
rachaelrenk wants to merge 2 commits into
mainfrom
docs/aeo-crosslinks-agents-orchestration-2

Conversation

@rachaelrenk
Copy link
Copy Markdown
Contributor

AEO cross-link audit: agents, cloud agents, and orchestration

Goal

Identify small internal cross-link improvements for agents, cloud agents, and orchestration docs — the pilot topic area for the AEO cross-link audit skill.

Source signals

Google Search Console (March–June 2026):

  • "warp multi agent" (18 impressions, 4 clicks) → lands on guides and getting-started pages but not the orchestration docs
  • "cloud orchestration" / "warp orchestration" (11+ impressions) → lands on the Oz Platform page and homepage, not the orchestration page
  • "ai coding agent orchestration" → lands on agent-platform/ root

These queries indicate users searching for orchestration and multi-agent concepts are landing on pages that don't link to the orchestration docs.

Existing-docs signals:

  • planning.mdx documents /plan, which is one of the two entry points for orchestration, but didn't link to orchestration
  • agents-in-warp.mdx mentions "Execute in parallel across repos or tasks" without explaining how
  • harnesses/index.mdx lists shared platform features but omitted orchestration, even though orchestration supports mixed harnesses
  • environments.mdx describes the execution flow without mentioning how orchestration uses environments

Peec: Not available in this environment. Recommendations are grounded in GSC + repo signals only.

Pages touched

File Why
capabilities/planning.mdx /plan proposes orchestration when beneficial — readers learning about planning should discover this
getting-started/agents-in-warp.mdx "Parallel execution" bullet needed a path for the reader to learn how
cloud-agents/harnesses/index.mdx Orchestration works across harnesses — readers evaluating harnesses should know this
cloud-agents/environments.mdx Orchestrated child agents each get their own environment — readers configuring environments should understand this use case

Links added

Source page Target page Reader next step
planning.mdx → Next steps /agent-platform/cloud-agents/orchestration/ A reader who just learned about /plan wants to know they can use it to coordinate multiple agents
agents-in-warp.mdx → From local to cloud /agent-platform/cloud-agents/orchestration/ A reader exploring cloud agents wants to know how to run tasks in parallel
harnesses/index.mdx → What stays the same /agent-platform/cloud-agents/orchestration/#where-parent-and-child-agents-can-run A reader choosing a harness wants to know that a parent on Warp Agent can spawn children on Claude Code or Codex
environments.mdx → Execution flow /agent-platform/cloud-agents/orchestration/ A reader configuring environments wants to understand that orchestrated children each get their own environment

Open questions for human review

  • The harnesses link targets the #where-parent-and-child-agents-can-run anchor — verify this renders correctly in the live Astro build.
  • The environments edit adds a clause to an existing numbered-list item. Confirm the sentence reads naturally in context.

Conversation: https://app.warp.dev/conversation/3b0029f0-d95d-4f95-8fc6-8b13c681ed8f
Run: https://oz.warp.dev/runs/019e9702-c000-77a0-b913-10c63bc02ce1

This PR was generated with Oz.

Add 4 internal cross-links connecting agent platform pages to the
orchestration docs:

- planning.mdx: link to orchestration from Next Steps (the /plan command
  proposes orchestration when the task benefits)
- agents-in-warp.mdx: link to orchestration in the From local to cloud
  section where parallel execution is mentioned
- harnesses/index.mdx: add Orchestration to the What stays the same
  list (parent and child can run different harnesses)
- environments.mdx: note that orchestrated child agents each get their
  own environment in the execution flow

Co-Authored-By: Oz <oz-agent@warp.dev>
@cla-bot cla-bot Bot added the cla-signed label Jun 5, 2026
@vercel
Copy link
Copy Markdown

vercel Bot commented Jun 5, 2026

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

Project Deployment Actions Updated (UTC)
docs Ready Ready Preview, Comment Jun 5, 2026 5:33pm

Request Review

@rachaelrenk rachaelrenk marked this pull request as ready for review June 5, 2026 17:31
@oz-for-oss
Copy link
Copy Markdown
Contributor

oz-for-oss Bot commented Jun 5, 2026

@rachaelrenk

I'm starting a first review of this pull request.

You can view the conversation on Warp.

I completed the review and no human review was requested for this pull request.

Comment /oz-review on this pull request to retrigger a review (up to 3 times on the same pull request).

Powered by Oz

Copy link
Copy Markdown
Contributor

@oz-for-oss oz-for-oss Bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Overview

This PR adds targeted internal cross-links from planning, agents overview, harnesses, and environments pages to the multi-agent orchestration docs. The links are directionally useful for AEO and reader navigation, but one added sentence overstates how orchestrated child environments work and conflicts with the orchestration page's cloud-local execution model.

Concerns

  • The environments page now says each child agent gets its own environment, but orchestrations can also run children inside the parent's cloud environment when shared state is needed.
  • The planning cross-link would be more accurate if it made the user approval step explicit before child agents launch.
  • No security findings. No approved spec context was provided for implementation drift review.

Verdict

Found: 0 critical, 1 important, 1 suggestions

Request changes

Comment /oz-review on this pull request to retrigger a review (up to 3 times on the same pull request).

Powered by Oz

3. **Environment** – The task uses an environment to define execution context
4. **Host** – The environment runs on a host (Warp-hosted or self-hosted infrastructure).
5. **Agent execution** – The workflow runs in the prepared environment
5. **Agent execution** – The workflow runs in the prepared environment. In [orchestrated runs](/agent-platform/cloud-agents/orchestration/), each child agent gets its own environment so that parallel agents are isolated from each other.
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ [IMPORTANT] This contradicts the orchestration docs' Cloud → cloud-local mode, where children run inside the parent's cloud environment instead of getting separate environments.

Suggested change
5. **Agent execution** – The workflow runs in the prepared environment. In [orchestrated runs](/agent-platform/cloud-agents/orchestration/), each child agent gets its own environment so that parallel agents are isolated from each other.
5. **Agent execution** – The workflow runs in the prepared environment. In [orchestrated cloud runs](/agent-platform/cloud-agents/orchestration/), child agents can run in separate environments for isolation or inside the parent's cloud environment when they need shared state.


* **[Interactive Code Review](/agent-platform/local-agents/interactive-code-review/)** - Leave inline comments on agent-generated diffs and have the agent revise in one pass.
* **[Cloud Agents quickstart](/agent-platform/cloud-agents/quickstart/)** - Run agents in the cloud for longer tasks, background automation, or parallel work across repos.
* **[Multi-agent orchestration](/agent-platform/cloud-agents/orchestration/)** - When `/plan` detects that a task benefits from parallel work, it proposes an orchestration config that spawns child agents to divide the work.
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

💡 [SUGGESTION] Clarify that /plan proposes orchestration for approval before child agents launch.

Suggested change
* **[Multi-agent orchestration](/agent-platform/cloud-agents/orchestration/)** - When `/plan` detects that a task benefits from parallel work, it proposes an orchestration config that spawns child agents to divide the work.
* **[Multi-agent orchestration](/agent-platform/cloud-agents/orchestration/)** - When a `/plan` task benefits from parallel work, the agent can propose an orchestration config for you to approve before child agents divide the work.

@rachaelrenk
Copy link
Copy Markdown
Contributor Author

First test run; made some tweaks to skill used by scheduled agent so closing this PR in favor of next week's run

@rachaelrenk rachaelrenk closed this Jun 5, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant