Crate agent_block_testkit 

In-process mock LLM servers and a provider wire-shape catalog for testing tool-calling clients.

Two pieces:

• shapes — builders for provider response bodies. Alongside the spec-conformant shapes, the catalog carries deliberately broken variants observed in the wild on OpenAI-compatible stacks (object arguments, missing id, malformed argument strings). A client that is only exercised against spec-shaped input silently loses tolerance for the shapes real serving stacks emit; the broken variants exist to guard exactly that failure mode.
• server — a small axum-backed mock server (server::MockLlm) that serves an Anthropic Messages (/v1/messages) or OpenAI Chat Completions (/chat/completions) endpoint from a caller-supplied responder closure, while recording request facts (tool declarations, tool_call_ids of returned tool results) for assertions.

use agent_block_testkit::server::MockLlm;
use agent_block_testkit::shapes::openai;
use serde_json::json;

let handle = MockLlm::openai(|req| {
    if !req.has_tool_results {
        // Broken-but-observed shape: object arguments, no id.
        openai::tool_calls_response(vec![openai::tool_call_object_args_no_id(
            "apply_search_replace",
            json!({"path": "/tmp/f.lua", "search": "a", "replace": "b"}),
        )])
    } else {
        openai::text_response("DONE")
    }
})
.spawn()
.await;
// point the client under test at handle.base_url ...
assert_eq!(handle.state.call_count(), 0);
handle.ct.cancel();

Modules

server

Axum-backed in-process mock LLM server.

shapes

Provider response-shape catalog: spec-conformant and deliberately broken tool-calling shapes.