docs(platform): refresh manual guides and links

[ntindle]

Why / What / How

Several manually authored guides had broken relative links after documentation assets moved, stale model lists after catalog changes, outdated ChatGPT connection instructions, and an incomplete External API scope table. This PR updates those guides against the current dev source, catalog, routes, and UI tests without touching generator-owned integration output.

Changes 🏗️

• Repair 19 AI/ML API and Ollama image references plus the legacy docs landing-page link.
• Point the External API guide at the existing OAuth guide and document WRITE_GRAPH and WRITE_LIBRARY.
• Align AI/ML API, Gemini, and Ollama model lists with the current LLM catalog.
• Align the ChatGPT/Codex preview steps with the current OpenAI integration labels, visible model tiers, and server-selected execution path.

Agents and large language models used

• Codex with GPT-5

Checklist 📋

For code changes:

• I have clearly listed my changes in the PR description
• I have made a test plan
• I have tested my changes according to the test plan:
  • poetry run python scripts/generate_block_docs.py --check
  • Full local Markdown target, exact-case, anchor, fence, and GitBook-directive scan
  • bash autogpt_platform/installer/test-single-container-handoff.sh
  • backend/copilot/codex_model_router_test.py (8 passed)
  • git diff --check origin/dev...HEAD
  • Repository pre-commit hooks for origin/dev...HEAD
  • backend/copilot/provider_tiers_test.py (the local SpinTestServer fixture stalled and was stopped after three minutes; no assertion failure was emitted)

Example test plan

• Create from scratch and execute an agent with at least 3 blocks
• Import an agent from file upload, and confirm it executes correctly
• Upload agent to marketplace
• Import an agent from marketplace and confirm it executes correctly
• Edit an agent from monitor, and confirm it executes correctly

For configuration changes:

• .env.default is updated or already compatible with my changes
• docker-compose.yml is updated or already compatible with my changes
• I have included a list of my configuration changes in the PR description (under Changes)

Examples of configuration changes

• Changing ports
• Adding new services that need to communicate with each other
• Secrets or environment variable changes
• New or infrastructure changes such as databases

---

Note

Low Risk
Markdown and link fixes only; no runtime, auth, or API behavior changes.

Overview
Documentation-only refresh for platform guides that had broken assets, stale catalog text, and outdated integration UI wording.

Relative links and screenshots now point at ../content/imgs/... (AI/ML API, Ollama) and the docs landing page uses ../platform/new_blocks.md. The External API guide links to oauth-guide.md instead of the removed SSO guide, and documents WRITE_GRAPH and WRITE_LIBRARY scopes.

Model tables in the AI/ML API, Gemini, and Ollama guides match the current LLM catalog (including a slimmer AI/ML API list and updated Gemini IDs); Ollama adds llama3.3 to the default compatible list.

The ChatGPT/Codex subscription preview doc matches current UI: connect via Settings → Integrations → OpenAI → ChatGPT → Sign in with ChatGPT, AutoPilot tier controls are Balanced/Advanced, and Codex routing is explained with an execution-path × tier mapping table instead of the older Fast/Thinking prose.

^{Reviewed by Cursor Bugbot for commit 0d678fe. Bugbot is set up for automated code reviews on this repo. Configure here.}

[cursor]

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: serverGenReqId_d14e6563-6790-4592-ae76-c414716069f8)

[github-actions]

🔍 PR Overlap Detection

This check compares your PR against all other open PRs targeting the same branch to detect potential merge conflicts early.

🔴 Merge Conflicts Detected

The following PRs have been tested and will have merge conflicts if merged after this PR. Consider coordinating with the authors.

• docs: fix links to removed or renamed files (FORGE-QUICKSTART, sso-guide) #14344 (simpleqt · updated 3d ago)

  • autogpt_platform/backend/backend/data/org_credit_test.py (1 conflict, ~10 lines)
  • docs/platform/integrating/api-guide.md (2 conflicts, ~10 lines)

• docs(platform): fix broken screenshot paths (../imgs -> ../content/imgs) #14345 (simpleqt · updated 3d ago)

  • 📁 autogpt_platform/backend/backend/data/
    • org_credit_test.py (1 conflict, ~10 lines)

• docs: remove broken waitlist link from docs landing page #14182 (putramkti · updated 5d ago)

  • 📁 docs/content/
    • index.md (1 conflict, ~6 lines)

Summary: 3 conflict(s), 0 medium risk, 0 low risk (out of 3 PRs with file overlap)

---

Auto-generated on push. Ignores: openapi.json, lock files.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Advanced

Run ID: 6d691f5a-608d-40e0-80a1-4c9caec37870

📥 Commits

Reviewing files that changed from the base of the PR and between c3542d1 and 0d678fe.

📒 Files selected for processing (2)

• docs/platform/codex-subscription-preview.md
• docs/platform/integrating/api-guide.md

Included review availability: Your plan provides up to 10 included reviews per hour; 8 remain after this review.

📜 Recent review details

⏰ Context from checks skipped due to timeout. (4)

• GitHub Check: Analyze (typescript)
• GitHub Check: Analyze (python)
• GitHub Check: Check PR Status
• GitHub Check: check-overlaps

🔇 Additional comments (2)

docs/platform/codex-subscription-preview.md (1)

75-75: LGTM!

Also applies to: 77-86

docs/platform/integrating/api-guide.md (1)

40-40: LGTM!

Also applies to: 51-51, 55-55, 74-74

---

Walkthrough

Documentation links and image paths were corrected. Provider model listings and Codex mappings were refreshed. The API guide now documents OAuth links and scopes.

Changes

Documentation updates

Layer / File(s)	Summary
Navigation and asset paths docs/content/index.md, docs/platform/aimlapi.md, docs/platform/ollama.md	Updated the Blocks link and documentation image paths. Added llama3.3 to the compatible Ollama models.
Provider model guidance docs/platform/aimlapi.md, docs/platform/codex-subscription-preview.md, docs/platform/gemini.md, docs/platform/ollama.md	Updated AI/ML API and Gemini model listings, Codex execution-path mappings, Gemini rate-limiting guidance, and Ollama model guidance.
OAuth guide updates docs/platform/integrating/api-guide.md	Replaced SSO links with OAuth links and added WRITE_GRAPH and WRITE_LIBRARY scopes.

Priority: ⬇️ Low — Defer this documentation-only update because it refreshes links, model catalogs, integration guidance, and OAuth scope descriptions without changing product behavior.

Estimated code review effort: 1 (Trivial) | ~5 minutes

Merge Risk: ⚪ Minimal · up to 0d678

This update refreshes documentation links, assets, provider guidance, and OAuth scope descriptions. The documented navigation and integration guidance are consistent with their referenced guides, with no remaining merge-readiness risk identified.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly summarizes the documentation refresh, including updated platform guides and links.
Description check	✅ Passed	The description directly explains the documentation link repairs, model updates, integration changes, and validation performed.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff. Analyzed 0 functions across 0…
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch codex/docs-code-audit-2026-09-08

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[ntindle]

/review

[autogpt-pr-reviewer]

🤖 Review of c3542d1 posted: #14447 (review)

[autogpt-pr-reviewer]

✅ Verdict: Approved — No findings require changes before approval; this review counts as an approval.

Documentation-only refresh of six Markdown guides; all image paths, relative links, OAuth scopes, and model catalog IDs verified accurate against current dev source by every specialist. GitHub CI is green on head c3542d1 and there are no code, API, or schema changes. No blockers.

Risk level: low | Human review: not required | Duration: 420s | Reviewed: c3542d19

Findings: 🔴 0 blockers | 🟠 0 should fix | 🟡 1 nice to have | 🔵 1 nit

Optional advice — does not block approval

🟡 Nice to Have

• 🔵 docs/platform/codex-subscription-preview.md Consider a table for the Codex tier mapping — The Balanced/Advanced tier-to-model routing prose is dense and harder to maintain over time.
  Suggestion: Optionally restructure the tier mapping as a small two-row table.

Optional advice — does not block approval

🔵 Nits

• 🔵 docs/platform/integrating/api-guide.md:51 Align WRITE_GRAPH description with enum semantics — WRITE_GRAPH is described as 'Create agent graphs', but the backend scope enum documents it as 'Can create and update agent graphs' (schema.prisma:2050), understating the permission.
  Suggestion: Change the description to 'Create and update agent graphs'.

GitHub CI on reviewed head: success

Validation and specialist details

Specialist Reports

Specialist	Status	Summary
security	✅ PASS	Documentation-only refresh of links, image paths, model lists, and OAuth scope docs; all links/scopes verified against current source with no security impact.
architect	✅ PASS	Documentation-only refresh with all image paths, links, OAuth scopes, and model IDs verified accurate against current dev source.
performance	✅ PASS	Documentation-only Markdown changes (link repaths, model tables, OAuth scope docs) with no runtime, query, or scalability impact.
testing	✅ PASS	Documentation-only PR; all relocated links, image paths, model IDs, and OAuth scopes verify against current dev source, with one cosmetic scope-description nit.
quality	✅ PASS	Documentation-only refresh whose links, model tables, and OAuth scopes all verify correctly against the current dev source and LLM catalog.
product	✅ PASS	Docs-only refresh; all image paths, links, OAuth scopes, and model catalogs verified accurate against current source with no user-facing defects.
discussion	✅ PASS	All required CI checks pass on head c3542d1 with no base merge conflict, and there are no open change requests from human reviewers.
ui-reviewer (local)	✅ PASS	Documentation-only PR; all image paths, relative links, OAuth scopes, and model IDs verified correct against current dev source with no defects introduced.
ui-reviewer (hosted)	✅ PASS	Documentation-only refresh whose repaired links all resolve and whose scope, model-catalog, and Codex routing claims all match current source; no defect introduced.

[cursor]

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: serverGenReqId_06b9d14b-9225-496c-8a2a-652e9e9b4b40)

[ntindle]

/reapprove

[gaoanze888]

Verified the link fixes actually resolve — the part a bot can't easily check.

Verified (relative paths resolve in the repo):

• ../imgs/aimlapi/ → ../content/imgs/aimlapi/: all 12 referenced Step PNGs exist at the new path; the old docs/imgs/ dir is gone, so the old links were genuinely broken. ✅
• ../imgs/ollama/ → ../content/imgs/ollama/: 9 images resolve at the new path. ✅
• content/index.md links: [Build your own Blocks](../platform/new_blocks.md) → docs/platform/new_blocks.md exists; [Contribute](contribute/index.md) → docs/content/contribute/index.md exists. ✅
• gemini.md is text-only (no image refs), so the empty docs/content/imgs/gemini/ dir is harmless — no broken links there. ✅

Content refreshes look right: model tables trimmed to the current catalog (aimlapi → Llama 3.3; gemini → 3.1/3 replacing 2.5/2.0), and the Codex ChatGPT connection flow now matches the OpenAI/ChatGPT selector path with the model-tier mapping table. The "currently available" phrasing is a good call given how often these catalogs churn.

No blockers from me — nice cleanup of the asset-move fallout.