[Kiro MITM] Request processing failed: ${error.message}

Error message

[Kiro MITM] Request processing failed: ${error.message}

What it means

The Kiro MITM handler forwards requests to the router and re-encodes the OpenAI SSE response into AWS EventStream binary via pipeTransformedEventStream. Any failure (router fetch, OpenAI→Kiro conversion, EventStream encoding) is logged '[Kiro MITM] Request processing failed: <message>' and returned as a 500 JSON error to the Kiro client.

Source

Thrown at src/mitm/handlers/kiro.js:522

    const tools = extractTools(body);

    const openaiBody = {
      model: mappedModel,
      messages,
      stream: true,
      // Forward tools so Claude uses structured tool_calls instead of XML text fallback
      ...(tools.length > 0 && { tools, tool_choice: "auto" }),
    };

    // 3: Forward to 9router
    const routerRes = await fetchRouter(openaiBody, "/v1/chat/completions", req.headers);

    // 4 + 5: Re-encode response as AWS EventStream binary using standard pipeline
    const state = initKiroState(mappedModel);

    await pipeTransformedEventStream(routerRes, res, convertOpenAIToKiro, state);
  } catch (error) {
    err(`[Kiro MITM] Request processing failed: ${error.message}`);
    if (!res.headersSent) {
      res.writeHead(500, { "Content-Type": "application/json" });
    }
    res.end(JSON.stringify({ 
      error: { 
        message: error.message, 
        type: "mitm_error",
        handler: "kiro"
      } 
    }));
  }
}

// Detect AWS EventStream binary format
function isBinaryEventStream(buffer) {
  if (!buffer || buffer.length < 12) return false;
  // AWS EventStream signature: 
  // - First 4 bytes: total frame length (big-endian)

View on GitHub (pinned to 90b52e06ff)

Solutions

Confirm the 9Router gateway is running and reachable at the configured router port.
Read the logged failure message: encoding errors usually mean a Kiro client/protocol update — update 9Router.
Verify the requested model exists in the Kiro model map / provider config.
Retry; if it fails mid-stream repeatedly, capture the dump file and compare against a working request.

Example fix

// before: protocol mismatch
convertOpenAIToKiro throws on unknown event shape → Request processing failed
// after
npm update 9router  # pull updated Kiro EventStream translation

Defensive patterns

Strategy: try-catch

Validate before calling

const health = await fetch('http://127.0.0.1:20128/dashboard').then(r => r.ok).catch(() => false);
if (!health) throw new Error('Router gateway not reachable');
if (!modelMap[kiroModel]) throw new Error(`No model mapping for ${kiroModel}`);

Try / catch

try {
  const state = initKiroState(mappedModel);
  await pipeTransformedEventStream(routerRes, res, convertOpenAIToKiro, state);
} catch (error) {
  err(`[Kiro MITM] Request processing failed: ${error.message}`);
  if (!res.headersSent) res.writeHead(500, { 'Content-Type': 'application/json' });
  res.end(JSON.stringify({ error: { message: error.message, type: 'mitm_error' } }));
}

Prevention

Keep Kiro model mappings current with the Kiro IDE's requested model ids.
Update 9Router when Kiro changes its EventStream framing.
Validate gateway health before starting the MITM listener.
Retain dump files from failed requests for protocol debugging.

When it happens

Trigger: intercept() throws during JSON parsing of the Kiro binary request, router call fails, or convertOpenAIToKiro/EventStream re-encoding encounters an unexpected chunk shape mid-stream.

Common situations: Kiro IDE pointed at the MITM host while the gateway is down; Kiro client updated its EventStream framing; a model mapping missing for the requested Kiro model.

Related errors

AI-assisted analysis of decolua/9router@90b52e06ff (2026-08-30). Data as JSON: /api/errors/eae2bc9af1b525db. Report an issue: GitHub.