Reorganize Agent Observability docs into task-based sections - #39258
Open
gsvigruha wants to merge 18 commits into
Open
Reorganize Agent Observability docs into task-based sections#39258gsvigruha wants to merge 18 commits into
gsvigruha wants to merge 18 commits into
Conversation
Move llm_observability pages into instrument/, investigate/, improve/, and configure/ sections, add redirects for every moved page, and update the English side navigation to match. Page content is unchanged apart from link paths. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
github-actions
Bot
added
Architecture
Rename the investigate section index title from Monitoring to Investigate, remove the Sensitive Data Scanner nav entry that has no page behind it, and repoint two pre-existing links that pointed at nonexistent targets. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Surface instrument/agentic and its Python, Node.js, Java, and prompt management pages in the side navigation. They were previously reachable only by direct URL. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The agentic prompt management page is reached from the agentic instrumentation guide, and its nav label duplicated the Prompt Management row under Configure. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The existing references to these pages use .md URLs intended for coding agents, which leaves the HTML pages without a clickable path. Add a whatsnext list covering all four subpages. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Contributor
Contributor
|
Replace the placeholder with an overview line and links to the evaluations, prompt management, and automation rules subsections. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Replace the placeholder with an overview line and links to the playground, datasets, and experiments subsections. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Rename the section directory, nav label, and identifier, then replace the placeholder with an overview line, a Get started code block, and links to the two subpages. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Inline the Homebrew install and authentication commands, and link to the Pup CLI documentation for the remaining installation methods. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Split the Get started section into a shared skills install and an either/or backend choice, and reword the intro. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The Build with AI landing page links both subpages, so the nav row is redundant. Align the data privacy page title and nav label with its filename. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The left nav template renders four levels and hides level-4 lists unless the parent identifier is allowlisted, which hid the evaluations and experiments children after they moved a level deeper. Allowlist the new parents, let the Agent Observability guides list expand, and add nav entries for the seven guides. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Remove the eight nav entries that sat at level 5, which the left nav template cannot render, and link the external evaluations subpages from their index instead. Keep the Evaluation Developer Guide row under Guides only. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Also rename the external evaluations heading to frameworks. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What does this PR do? What is the motivation?
Reorganizes the Agent Observability documentation into task-based sections. Existing pages are moved rather than rewritten; the new prose is limited to three section landing pages, a few link fixes, and title alignment.
New structure:
quickstart/(withterms/nested under it)build_with_ai/— MCP server and Claude Code skills pagesinstrument/— formerinstrumentation/, plusagent_observability_and_apmandprompt_trackinginvestigate/— querying, cost, metrics, patterns, annotation queues, export APIimprove/— playground, datasets, andexperiments/configure/—evaluations/, prompt management, automation rulesguide/— gains proxy services, agent monitoring, MCP client, and an AWS Lambda guideMoves and redirects:
git mv, so file history is preserved.content/en.Side navigation template:
layouts/partials/nav/left-nav.htmlrenders four levels below a product and hides level-4 lists unless the parent identifier is allowlisted. Moving Evaluations and Experiments one level deeper pushed their children into that gated level, so their submenus disappeared. Two scoped changes fix it:llm_obs_evaluations,llm_obs_experiments, andllm_obs_instrumentation_agenticto the level-4 allowlist.(ne .Identifier "llm_obs_guide"), so only Agent Observability's guides expand. No other product defines guide children, so nothing changes for them.Because the template stops at four levels, the children of
llm_as_a_judge_evaluationsandexternal_evaluationsare not in the nav. They are linked from their section index pages instead.Navigation and reachability:
instrument/agenticand its language pages, and for the seven guides.instrument/agentic/_index.mdandconfigure/evaluations/external_evaluations/_index.md. The existing references on the agentic page use.mdURLs intended for coding agents, which left the HTML pages without a clickable path.Sensitive Data Scannernav entry, which had no page behind it. Its URL still redirects through an alias.instrument/, the slug, nav label, and page title now line up on every page.New pages:
improve/_index.md,configure/_index.md, andbuild_with_ai/_index.mdare section landing pages with an overview line and links to their subsections.build_with_ai/_index.mdalso has a Get started section presenting the Datadog MCP server and the Pup CLI as alternative backends for the skills; every command in it is taken from an existing page.guide/trace-an-llm-application-in-aws-lambda.mdholds the AWS Lambda setup block, moved verbatim out ofinstrument/sdk.md, which now links to it. This replaces a dead anchor link on the guides index.Other fixes:
MonitoringtoInvestigate, and the data privacy page toData Privacy, Security, and RBACto match its filename.evaluation_typespage and aquickstart#trace-an-llm-applicationanchor.Known follow-ups, deliberately left out of this PR:
llm_observabilityURL segment toagent_observability. Doable with aliases, but it changes the public URL of an entire product and also requires updating theunsupported_siteskey inconfig/_default/params.yaml, which is looked up by path segment, plusdata/products.yamland thedata/partials/home*.yamlfiles. Planned as a separate PR.static/images/llm_observability/keep their old subdirectory names, because translated pages reference the same paths.fr,es,ja,ko) and the translated side navigations still use the old paths. Those pages keep building at their old URLs, but moved English pages have no localized counterpart until the translation pipeline catches up.guide/now inherit the algoliacascadeinguide/_index.md(Guide category, rank 20), andclaude_code_skillsno longer does.multi-modality,agent_observability_and_logs, andinsightspages do not exist yet.Merge readiness
For Datadog employees:
<name>/<description>convention and include the forward slash (/). If you've already created your PR with an incorrect branch name, please rename your branch and open a fresh PR./reviewto run an automated check that catches common issues before a Documentation team member reviews your PR.AI assistance
Claude Code performed the moves, alias and link updates, and navigation rewrite from an approved mapping, and drafted the new section landing pages.
Additional notes
Redirects were validated statically (every internal
llm_observabilityreference resolves to a real page or alias, apart from.mdURLs that are intentionally aimed at coding agents) rather than against a local Hugo build. Worth running htmltest before merge.