Playbook: Knowledge maintenance

Use this playbook for recurring knowledge-base maintenance, failed-ingestion cleanup, source updates, and content quality checks.

Required access

Recommended scopes:

sitegpt login \
  --profile knowledge-agent \
  --scope account:read \
  --scope chatbots:read \
  --scope knowledge:read \
  --scope knowledge:write \
  --scope knowledge:delete

Add settings:read if the agent should inspect chatbot settings and settings:write if it should update crawler-related settings.

1. Inspect health

sitegpt dashboard --chatbot <chatbot-id> --json
sitegpt knowledge documents stats --chatbot <chatbot-id> --json
sitegpt knowledge documents list --chatbot <chatbot-id> --status FAILED --json
sitegpt knowledge documents list --chatbot <chatbot-id> --status PROCESSING --json
sitegpt knowledge sync-jobs list --chatbot <chatbot-id> --json

Look for:

Failed documents
Stale source documents
Missing sitemap sync/scan jobs
Duplicated documents
Outdated custom responses
Large pending queues

2. Resync failed or stale documents

Resync all failed documents:

sitegpt knowledge documents resync --chatbot <chatbot-id> --state failed

Resync by source:

sitegpt knowledge documents resync --chatbot <chatbot-id> --source WEBSITE
sitegpt knowledge documents resync --chatbot <chatbot-id> --source SITEMAP

Resync exact documents:

sitegpt knowledge documents resync --chatbot <chatbot-id> --document <document-id> --document <document-id>

3. Update document config

For pages with layout noise:

sitegpt knowledge documents update-config \
  --chatbot <chatbot-id> \
  --document <document-id> \
  --only-main-content true \
  --include-selector main \
  --exclude-selector nav \
  --exclude-selector footer

For recurring content refresh:

sitegpt knowledge documents update-config --chatbot <chatbot-id> --document <document-id> --sync WEEKLY

Valid sync frequencies:

NEVER, DAILY, WEEKLY, MONTHLY

4. Maintain sync jobs

List active jobs:

sitegpt knowledge sync-jobs list --chatbot <chatbot-id>

Update job frequency:

sitegpt knowledge sync-jobs update --chatbot <chatbot-id> <job-id> --sync WEEKLY
sitegpt knowledge sync-jobs update --chatbot <chatbot-id> <job-id> --scan WEEKLY

Disable a job:

sitegpt knowledge sync-jobs delete --chatbot <chatbot-id> <job-id> --yes

Sitemap --scan keeps sitemap URL additions/deletions up to date. --sync refreshes content for known URLs.

5. Add new website content

Prefer sitemap:

sitegpt knowledge sitemap add \
  --chatbot <chatbot-id> \
  https://example.com/sitemap.xml \
  --max-links 500 \
  --only-main-content true \
  --sync WEEKLY \
  --scan WEEKLY

Fallback to website crawl:

sitegpt knowledge website add \
  --chatbot <chatbot-id> \
  https://example.com/docs \
  --depth 3 \
  --max-links 200 \
  --only-main-content true

Use exact links for surgical updates:

sitegpt knowledge links add \
  --chatbot <chatbot-id> \
  https://example.com/pricing \
  https://example.com/refund-policy \
  --only-main-content true

6. Maintain source connections

List sources:

sitegpt knowledge sources list --chatbot <chatbot-id> --json

Select more files for an existing picker connection:

sitegpt knowledge sources authorize --chatbot <chatbot-id> <source-id>
sitegpt knowledge sources documents --chatbot <chatbot-id> <source-id>
sitegpt knowledge sources ingest --chatbot <chatbot-id> <source-id>

Ingest selected documents only:

sitegpt knowledge sources ingest \
  --chatbot <chatbot-id> \
  <source-id> \
  --external-id <external-document-id> \
  --sync WEEKLY

GitHub:

sitegpt knowledge sources github repos --chatbot <chatbot-id> --source <source-id>
sitegpt knowledge sources github files --chatbot <chatbot-id> --source <source-id> --owner <owner> --repo <repo>
sitegpt knowledge sources ingest --chatbot <chatbot-id> <source-id> --repo <repo> --branch main --pattern "docs/**"

Confluence:

sitegpt knowledge sources confluence spaces --chatbot <chatbot-id> --source <source-id>
sitegpt knowledge sources confluence pages --chatbot <chatbot-id> --source <source-id> --space <space-id>
sitegpt knowledge sources ingest --chatbot <chatbot-id> <source-id> --page <page-id>

7. Add or update text knowledge

Use text knowledge for short curated policy text, summaries, or controlled notes:

sitegpt knowledge text update --chatbot <chatbot-id> --name "Support policy notes" --file ./support-notes.md

Because this replaces the text document, read existing content first when editing:

sitegpt knowledge documents list --chatbot <chatbot-id> --source TEXT --json
sitegpt knowledge documents get --chatbot <chatbot-id> <document-id> --content --json

8. Maintain Custom Responses

List open Custom Responses:

sitegpt knowledge custom-responses list --chatbot <chatbot-id> --state OPEN --json

Add a new approved answer:

sitegpt knowledge custom-responses add \
  --chatbot <chatbot-id> \
  --question "Do you offer refunds?" \
  --answer "Refund eligibility depends on your plan and billing terms. Contact support@example.com for account-specific help."

Update an existing response:

sitegpt knowledge custom-responses update --chatbot <chatbot-id> <custom-response-id> --answer "Updated approved answer."

Delete outdated responses:

sitegpt knowledge custom-responses delete --chatbot <chatbot-id> <custom-response-id> --yes

9. Delete obsolete documents carefully

Preview matching documents:

sitegpt knowledge documents list --chatbot <chatbot-id> --query "old product" --json

Delete exact IDs:

sitegpt knowledge documents delete --chatbot <chatbot-id> --document <document-id> --document <document-id> --yes

Delete failed documents only:

sitegpt knowledge documents delete --chatbot <chatbot-id> --state failed --yes

Avoid --all --yes unless the user explicitly asked for broad deletion.

10. Final verification

sitegpt dashboard --chatbot <chatbot-id> --json
sitegpt knowledge documents stats --chatbot <chatbot-id> --json
sitegpt knowledge documents list --chatbot <chatbot-id> --status FAILED --json
sitegpt messages send --chatbot <chatbot-id> "What topics are you trained on?" --json

Report:

What was added or updated
Any failed documents
Any sync jobs changed
Any deleted documents or custom responses
Follow-up work requiring human approval

​Required access

​1. Inspect health

​2. Resync failed or stale documents

​3. Update document config

​4. Maintain sync jobs

​5. Add new website content

​6. Maintain source connections

​7. Add or update text knowledge

​8. Maintain Custom Responses

​9. Delete obsolete documents carefully

​10. Final verification