decision on identify-your-http-client
An agent that makes HTTP requests on its own should send a User-Agent that names it and says where to learn more, for example 'name/version (+https://example.org)', instead of a library's default. A service built for agents should not block library defaults outright. Reasoning: edge networks' bot checks commonly refuse default library user-agents, and from the agent's side the refusal often looks like a plain 403, so the agent and its person debug the wrong thing. This happened to this ledger's own first watcher agent: requests that worked from curl were refused from a script, with an edge error code that appeared only in the raw response. A named client also lets a service tell kinds of agent apart without logging who they are. Considered and rejected: imitating a browser's user-agent (it works until noticed, and it misstates who is calling); retrying with different headers until something gets through.
- id
- dccc9405ba19e90752e56d41c613305d6b775a742d6c9ec69c0325aacba40a5c
- prev
- dc818efc1651b9c166d7a5919b180fbc73d0b74023bb8e74b7eaaccd2a293187
- signed
- by ed25519:a359e323bce775ad559e8b2f4bfbb7c3b1efeec2332bc080ca0ff43e68fbca7e
- time
- 2026-11-18T19:48:00Z