AWS Bedrock Claude Unusually Large Prompts

name: AWS Bedrock Claude Unusually Large Prompts
id: 95fbdd4e-1202-4971-b85a-513b2005086e
version: 1
author: Rod Soto
status: production
type: Anomaly
creation_date: '2026-07-06'
modification_date: '2026-07-06'
security_domain: endpoint
data_source:
    - AWS Bedrock Claude
category: application
description: This detection identifies unusually large prompts sent to AWS Bedrock Claude models by calculating the statistical baseline of input token counts and flagging requests that exceed 15 standard deviations above the mean. Abnormally large prompts may indicate prompt injection attacks, data exfiltration attempts, or abuse of the AI service.
search: |-
    `aws_bedrock_claude`
    | rename "input.inputTokenCount" AS input_tokens
    | rename "identity.arn" AS user_arn
    | rename "accountId" AS account_id
    | rex field=user_arn "assumed-role/[^/]+/(?<user>[^\"]+)$"
    | eval input_tokens=tonumber(input_tokens)
    | where input_tokens>0
    | eventstats avg(input_tokens) AS avg_tokens stdev(input_tokens) AS stdev
    | where input_tokens > avg_tokens + (1*stdev)
    | where input_tokens > 1000
    | table _time, modelId, user, user_arn, account_id, input_tokens, avg_tokens, host
    | sort - input_tokens
    | `aws_bedrock_claude_unusually_large_prompts_filter`
how_to_implement: You must install and configure the Splunk Add-on for AWS (https://splunkbase.splunk.com/app/1876). Enable Amazon Bedrock model invocation logging in AWS so that Claude request/response payloads are delivered to S3 and/or CloudWatch Logs (see https://docs.aws.amazon.com/bedrock/latest/userguide/model-invocation-logging.html for setup steps), then ingest those logs into Splunk via the AWS TA. Configure the `aws_bedrock_claude` macro to point to the index and sourcetype (`json_no_timestamp`) where these logs land.
known_false_positives: Possible false positives may arise from legitimate user interactions with the AI model that require large input prompts, such as complex queries or multi-turn conversations. It is important to review the context of the detected prompts to determine if they represent actual prompt injection attempts or benign usage.
references:
    - https://aws.amazon.com/blogs/apn/unlocking-the-power-of-splunk-with-amazon-bedrock-an-agentic-ai-approach-to-build-customized-splunk-assistants-using-bedrock-agents/
    - https://help.splunk.com/en/splunk-observability-cloud/observability-for-ai/splunk-ai-infrastructure-monitoring/set-up-ai-infrastructure-monitoring/amazon-bedrock
    - https://research.splunk.com/stories/aws_bedrock_security/
    - https://docs.aws.amazon.com/bedrock/latest/userguide/model-invocation-logging.html
drilldown_searches:
    - name: View the detection results for - "$host$"
      search: '%original_detection_search% | search host="$host$"'
      earliest_offset: $info_min_time$
      latest_offset: $info_max_time$
    - name: View risk events for the last 7 days for - "$host$"
      search: '| from datamodel Risk.All_Risk | search normalized_risk_object IN ("$host$") | stats count min(_time) as firstTime max(_time) as lastTime values(search_name) as "Search Name" values(risk_message) as "Risk Message" values(analyticstories) as "Analytic Stories" values(annotations._all) as "Annotations" values(annotations.mitre_attack.mitre_tactic) as "ATT&CK Tactics" by normalized_risk_object | `security_content_ctime(firstTime)` | `security_content_ctime(lastTime)`'
      earliest_offset: 7d
      latest_offset: "0"
intermediate_findings:
    entities:
        - field: host
          type: system
          score: 20
          message: Unusually large prompt detected on $host$ with $input_tokens$ input tokens, exceeding the statistical baseline of $avg_tokens$ average tokens.
analytic_story:
    - Suspicious AWS Bedrock Claude Activities
asset_type: Web Application
mitre_attack_id:
    - T1055
product:
    - Splunk Enterprise
    - Splunk Enterprise Security
    - Splunk Cloud
tests:
    - name: True Positive Test
      test_type: unit
      attack_data:
        - data: https://media.githubusercontent.com/media/splunk/attack_data/master/datasets/aws_bedrock_claude/aws_bedrock_claude_unusually_large_prompts.ndjson
          sourcetype: json_no_timestamp
          source: http:bulkawsbedrock