BotsRated

Policy

Set the rules your agents check before they pay or call a service. One policy per account.

Stays in this browser tab. Sent only as an Authorization header to the policy service; never saved by Desk.
Your choice; your software applies it.
One hostname per line, e.g. api.example.com. Lowers overridable rules to info.
One hostname per line. Always answers block.
Payee addresses and custom rules (advanced)
One 0x address per line. Stored; no effect in v0.
One 0x address per line. Public addresses only, never a private key.
Up to 50 rules, e.g. [{"id":"cap","when":{"max_amount_atomic":"500000"},"then":"block"}]. Conditions: payto_not_observed, payto_changed_within, price_above_observed, price_above_listing, max_amount_atomic, not_covered, data_stale, no_valid_402_within, valid_402_rate_below, delisted, network_or_asset_not, resource_url_mismatch, mcp_tools_changed_within, mcp_tools_hash_not, mcp_state_not_ok, test_buy_failed_recent.

Limits: 16 KB, 200 list entries in total. The policy service checks every save; nothing is stored if it refuses.

Try a check

See what your saved policy answers for one service.

Usage

Event totals for the last 30 days. Counts only; no balances.

Your own rules, applied to BotsRated measurements. Not a BotsRated rating of anyone. Desk doesn't run your agents and never holds or moves money; your own software decides what to do with each answer.