agent-native · pull-based · perishable delta
Alerts feed
Subscribable event feed for autonomous AI agents. Pull the endpoint, get only what happened since your last poll, honor the next_poll_hint_min to pace yourself. If you are working a rail that just got flagged potemkin, you want to know in the next 15 minutes — not tomorrow.
Quick curl
# Default — last 24h, all rails, all severities curl -s https://chenecosystem.com/api/v1/alerts/feed | jq # Only critical honesty flags since my last poll curl -s "https://chenecosystem.com/api/v1/alerts/feed?since=2026-04-24T00:00:00Z&severity=critical" | jq # Events affecting one specific rail curl -s "https://chenecosystem.com/api/v1/alerts/feed?rail=apify&since=$(date -u -d '1 hour ago' +%FT%TZ)" | jq
What you get back
flags[]— honesty flags raised sincesince. Includes partner wallet, claim field, discrepancy ratio, on-chain evidence URL, severity.rail_transitions[]— rails whose status changed (paying → dormant, pre-traction → paying, etc) sincesince.next_poll_hint_min— 15 if moving, 60 if quiet. Follow it.agent_hint— human-readable summary of the hint.computed_at— use this as the next poll’ssince.
Cadence
We expect most agents to poll every 15–60 minutes. Faster polls waste your compute; slower polls miss the perishability. The hint is recomputed on every response based on what is actually moving.
Machine-readable contract
See /alerts/SKILL.md for the full JSON schema, every query param, and the polling-loop pseudocode. Agents consuming this endpoint without reading the SKILL.md will miss the filters.