What causes "[Agent:${this.name}] - Input processor error" in mastra-ai/mastra?

Read `error.details.processorId` and `reason` to find the failing processor, then fix or harden that processor's code

mastra-ai/mastra · error · MastraError

AGENT_INPUT_PROCESSOR_ERROR

AGENT_INPUT_PROCESSOR_ERROR

Error message

[Agent:${this.name}] - Input processor error

What it means

A MastraError (id AGENT_INPUT_PROCESSOR_ERROR, category USER) thrown when an input processor attached to the agent throws during processing of the inbound request. The agent wraps the processor's own error with agent name, processorId, and retry details, attributing the failure to user-supplied processor logic rather than core library code.

Source

Thrown at packages/core/src/agent/agent.ts:4279

      });
      try {
        messageList = await runner.runInputProcessors(messageList, observabilityContext, requestContext, 0);
      } catch (error) {
        if (error instanceof TripWire) {
          tripwire = {
            reason: error.message,
            retry: error.options?.retry,
            metadata: error.options?.metadata,
            processorId: error.processorId,
          };
          this.logger.warn('Input processor tripwire triggered', {
            agent: this.name,
            reason: error.message,
            processorId: error.processorId,
            retry: error.options?.retry,
          });
        } else {
          throw new MastraError(
            {
              id: 'AGENT_INPUT_PROCESSOR_ERROR',
              domain: ErrorDomain.AGENT,
              category: ErrorCategory.USER,
              text: `[Agent:${this.name}] - Input processor error`,
            },
            error,
          );
        }
      }
    }

    return {
      messageList,
      tripwire,
    };
  }

View on GitHub (pinned to 75dd419e61)

Solutions

Read `error.details.processorId` and `reason` to find the failing processor, then fix or harden that processor's code
Add try/catch inside the processor and return a fallback result or throw a descriptive error with actionable `options` (e.g. retry)
If the processor is optional, remove it from the agent's inputProcessors config or make it no-op on failure
Check processor API compatibility with your @mastra/core version

Example fix

// before
const processor = { processInput: async ({ messages }) => callModerationApi(messages) };
// after
const processor = { processInput: async ({ messages }) => {
  try { return await callModerationApi(messages); }
  catch (e) { console.warn('moderation unavailable, skipping', e); return { messages };
} };

Defensive patterns

Strategy: try-catch

Type guard

function isInputProcessorError(e) {
  return e instanceof Error && 'id' in e && e.id === 'AGENT_INPUT_PROCESSOR_ERROR';
}

Try / catch

try {
  await agent.generate(prompt, opts);
} catch (e) {
  if (e?.id === 'AGENT_INPUT_PROCESSOR_ERROR') {
    logger.error(`processor ${e.details.processorId} failed: ${e.details.reason}`);
    return fallbackResponse();
  }
  throw e;
}

Prevention

Wrap external calls inside every input processor in try/catch with an explicit fallback
Log and alert on processor failures with processorId
Keep processors pure/cheap; push flaky network work behind retries
Test processors against malformed input before deploying

When it happens

Trigger: Any `agent.generate()`/`stream()`/`loop()` call where a registered InputProcessor's `processInput` (or similar) callback throws, and the failure is not recoverable via the processor's retry options.

Common situations: Custom input processors that call external services which are down; processors validating prompts that throw on malformed input; processors written against an older processor API shape; throws inside async processor code not caught by the processor itself.

Related errors

AI-assisted analysis of mastra-ai/mastra@75dd419e61 (2026-08-30). Data as JSON: /api/errors/13a591ed72a48dbe. Report an issue: GitHub.