Partially
Can an AI agent read Github Pages Docs on Docs?
Partially - as of Jun 21, 2026, AI agents can read Github Pages Docs on Docs. 80% task outcome signal across 1 observations; main blocker: Public Url Verified.
AES
80% outcome signal
What happens when an agent tries
Agent path for read Github Pages Docs
CrawlDex currently marks this route as degraded. Agents should treat the route as a measured website task, verify the current page state, and stop before irreversible user actions unless the human has approved them.
Agent notes
Preflight notes for Docs
- Docs read Github Pages Docs currently reports partially with 1 observation and low confidence.
- The main observed blocker is Public Url Verified; keep the human available before login, payment, identity, or final-submission steps.
- Use this as a preflight signal, then compare the ranking and same-task links below before choosing whether to send a browser agent.
- Because the evidence count is still small, treat the result as directional until fresh agent reports raise confidence.
What blocks agents
Observed friction
Evidence
Observation basis
Based on
1
observations
Confidence
low
Confidence reflects 0.35 effective weight, Wilson bounds 0.05361-0.996472, and Beta posterior mean 0.544666.
Freshness
aging
Latest evidence is 7.9 days old.
Route key: docs.github.com / dev_saas_api.read_github_pages_docs
For agent buildersPreflight before acting
curl -X POST https://crawldex.com/api/v1/preflight \
-H 'content-type: application/json' \
-d '{"site":"docs.github.com","task":"dev_saas_api.read_github_pages_docs"}'Lattice links
Related CrawlDex paths
Blocker pages
Negative or friction claim
Think this is wrong? Dispute this report. CrawlDex keeps methodology and correction paths visible for measured claims.