Back to catalog
Static reviewDevelopment

Skill dossier

detecting-business-email-compromise-with-ai

Source declaration

Deploy AI and NLP-powered detection systems to identify business email compromise attacks by analyzing writing style, behavioral patterns, and contextual anomalies that evade traditional rule-based filters.

The author's original description, preserved in English. It describes intent, not verified performance.

Deploy AI and NLP-powered detection systems to identify business email compromise attacks by analyzing writing style, behavioral patterns, and contextual anomalies that evade traditional rule-based filters.

developmentreferencesscripts

Research summary

Machine-generated analysis; pending human review

These observations come from deterministic source parsing, not a human security review or controlled task evaluation.

Deploy AI and NLP-powered detection systems to identify business email compromise attacks by analyzing writing style, behavioral patterns, and contextual anomalies that evade traditional rule-based filters. This bilingual summary is generated deterministically from collected source text and file structure; it is pending human editorial review.

Review scope

Deterministic static analysis of collected SKILL.md and supporting text/code. Machine-assisted candidates and findings await human editorial review; no dynamic behavior or effect is assessed.

Observable capabilities

  • Workflow [Source statement; provisional, not tested. Evidence: skills/detecting-business-email-compromise-with-ai/SKILL.md:L100]
  • Step 1: Deploy AI Email Security Platform [Source statement; provisional, not tested. Evidence: skills/detecting-business-email-compromise-with-ai/SKILL.md:L102]
  • Step 2: Configure Behavioral Baselines [Source statement; provisional, not tested. Evidence: skills/detecting-business-email-compromise-with-ai/SKILL.md:L109]
  • Step 3: Train NLP Models for BEC Detection [Source statement; provisional, not tested. Evidence: skills/detecting-business-email-compromise-with-ai/SKILL.md:L116]
  • Step 4: Configure Detection Policies [Source statement; provisional, not tested. Evidence: skills/detecting-business-email-compromise-with-ai/SKILL.md:L123]
  • Step 5: Integrate with Response Workflow [Source statement; provisional, not tested. Evidence: skills/detecting-business-email-compromise-with-ai/SKILL.md:L130]

Limitations & scope

  • Static analysis cannot assess runtime behavior, command success, network reachability, or model output quality.
  • No commands, scripts, network requests, model calls, or controlled skill executions were performed.
  • Category, tags, capabilities, workflows, and bilingual editorial summaries are deterministic candidates pending human review, not verified outcomes.
  • Exact token counts are unavailable until a later pipeline performs and records an official model-specific count.

Relevant for

  • Practitioners evaluating whether to inspect and install the source skill; this audience is inferred from the product context, not measured usage. [Evidence: skills/detecting-business-email-compromise-with-ai/SKILL.md:L1-L79]

What ships with the skill

The file inventory comes from the collected source at the recorded commit. Paths and sizes are structural observations, not a measure of quality.

files
8
Markdown files
5
scripts
2
references
4
Size
38.4 KB
Structural complexity
Low

File inventory

PathTypeBytes
skills/detecting-business-email-compromise-with-ai/assets/template.mdReference927 B
skills/detecting-business-email-compromise-with-ai/LICENSEOther11 KB
skills/detecting-business-email-compromise-with-ai/references/api-reference.mdReference2.2 KB
skills/detecting-business-email-compromise-with-ai/references/standards.mdReference1.3 KB
skills/detecting-business-email-compromise-with-ai/references/workflows.mdReference1.4 KB
skills/detecting-business-email-compromise-with-ai/scripts/agent.pyScript5.3 KB
skills/detecting-business-email-compromise-with-ai/scripts/process.pyScript10.8 KB
skills/detecting-business-email-compromise-with-ai/SKILL.mdInstruction5.6 KB

SKILL.md frontmatter

{
  "name": "detecting-business-email-compromise-with-ai",
  "description": "Deploy AI and NLP-powered detection systems to identify business email compromise attacks by analyzing writing style, behavioral patterns, and contextual anomalies that evade traditional rule-based filters.",
  "domain": "cybersecurity",
  "subdomain": "phishing-defense",
  "tags": [
    "bec",
    "ai",
    "nlp",
    "machine-learning",
    "email-security",
    "behavioral-analytics",
    "impersonation",
    "fraud-detection"
  ],
  "version": "1.0",
  "author": "mahipal",
  "license": "Apache-2.0",
  "atlas_techniques": [
    "AML.T0073",
    "AML.T0052",
    "AML.T0088"
  ],
  "nist_ai_rmf": [
    "GOVERN-6.2",
    "MAP-5.2",
    "GOVERN-6.1",
    "MEASURE-2.7",
    "MEASURE-2.5"
  ],
  "d3fend_techniques": [
    "Sender MTA Reputation Analysis",
    "Email Filtering",
    "Sender Reputation Analysis",
    "Homoglyph Detection",
    "Message Analysis"
  ],
  "nist_csf": [
    "PR.AT-01",
    "DE.CM-09",
    "RS.CO-02",
    "DE.AE-02"
  ],
  "mitre_attack": [
    "T1566.002",
    "T1534",
    "T1114.002",
    "T1657",
    "T1078.004"
  ],
  "mitre_f3": {
    "version": "1.1",
    "tactics": [
      "initial-access",
      "stealth",
      "positioning",
      "monetization"
    ],
    "techniques": [
      {
        "id": "T1660",
        "name": "Phishing",
        "tactic": "initial-access",
        "source": "attack"
      },
      {
        "id": "T1672",
        "name": "Email Spoofing",
        "tactic": "stealth",
        "source": "attack"
      },
      {
        "id": "F1032",
        "name": "Impersonate Official",
        "tactic": "initial-access",
        "source": "f3"
      },
      {
        "id": "F1005.006",
        "name": "Account Manipulation: Change of Payment Details",
        "tactic": "positioning",
        "source": "f3"
      },
      {
        "id": "F1022",
        "name": "Delete Relevant Emails",
        "tactic": "stealth",
        "source": "f3"
      },
      {
        "id": "F1025.003",
        "name": "Electronic Funds Transfer: Wire Transfer",
        "tactic": "monetization",
        "source": "f3"
      }
    ]
  }
}

A reading of the operating sequence

This workflow reconstructs the apparent sequence from source material. Unless controlled evaluation is explicitly recorded, it is provisional—not an observed trace.

  1. 01
    Action

    Overview

    Source-described action candidate under “Overview”; not verified execution.

  2. 02
    Decision

    When to Use

    Source-described decision candidate under “When to Use”; not verified execution.

  3. 03
    Action

    Workflow

    Source-described action candidate under “Workflow”; not verified execution.

  4. 04
    Action

    Step 1: Deploy AI Email Security Platform

    Source-described action candidate under “Step 1: Deploy AI Email Security Platform”; not verified execution.

  5. 05
    Action

    Select API-based solution (Abnormal Security, Tessian, Ironscales) or enhance existing SEG

    Source-described action candidate under “Step 1: Deploy AI Email Security Platform”; not verified execution.

  6. 06
    Action

    Connect to Microsoft Graph API or Google Workspace API

    Source-described action candidate under “Step 1: Deploy AI Email Security Platform”; not verified execution.

Inspect all extracted workflow candidates (35 candidate steps)
  1. 07
    Action

    Allow 48-hour baseline learning period on historical email data

    Source-described action candidate under “Step 1: Deploy AI Email Security Platform”; not verified execution.

  2. 08
    Action

    Configure integration to scan inbound, outbound, and internal email

    Source-described action candidate under “Step 1: Deploy AI Email Security Platform”; not verified execution.

  3. 09
    Action

    Verify API permissions for message access and remediation

    Source-described action candidate under “Step 1: Deploy AI Email Security Platform”; not verified execution.

  4. 10
    Action

    Step 2: Configure Behavioral Baselines

    Source-described action candidate under “Step 2: Configure Behavioral Baselines”; not verified execution.

  5. 11
    Action

    AI learns normal communication patterns: who emails whom, frequency, tone

    Source-described action candidate under “Step 2: Configure Behavioral Baselines”; not verified execution.

  6. 12
    Action

    Establish writing style profiles for each user (vocabulary, sentence structure)

    Source-described action candidate under “Step 2: Configure Behavioral Baselines”; not verified execution.

  7. 13
    Action

    Map typical request types per role (finance processes payments, HR handles PII)

    Source-described action candidate under “Step 2: Configure Behavioral Baselines”; not verified execution.

  8. 14
    Action

    Baseline email metadata: typical sending times, devices, locations

    Source-described action candidate under “Step 2: Configure Behavioral Baselines”; not verified execution.

  9. 15
    Action

    Flag deviations from established baselines as anomalous

    Source-described action candidate under “Step 2: Configure Behavioral Baselines”; not verified execution.

  10. 16
    Action

    Step 3: Train NLP Models for BEC Detection

    Source-described action candidate under “Step 3: Train NLP Models for BEC Detection”; not verified execution.

  11. 17
    Action

    Deploy transformer-based models (BERT, GPT) for email content analysis

    Source-described action candidate under “Step 3: Train NLP Models for BEC Detection”; not verified execution.

  12. 18
    Action

    Detect urgency and manipulation language patterns

    Source-described action candidate under “Step 3: Train NLP Models for BEC Detection”; not verified execution.

  13. 19
    Action

    Identify mismatches between sender identity and writing style

    Source-described action candidate under “Step 3: Train NLP Models for BEC Detection”; not verified execution.

  14. 20
    Action

    Analyze sentiment shifts indicating social engineering pressure

    Source-described action candidate under “Step 3: Train NLP Models for BEC Detection”; not verified execution.

  15. 21
    Action

    Classify email intent: information request, payment request, credential request

    Source-described action candidate under “Step 3: Train NLP Models for BEC Detection”; not verified execution.

  16. 22
    Action

    Step 4: Configure Detection Policies

    Source-described action candidate under “Step 4: Configure Detection Policies”; not verified execution.

  17. 23
    Action

    VIP impersonation: AI compares new email against known executive communication patterns

    Source-described action candidate under “Step 4: Configure Detection Policies”; not verified execution.

  18. 24
    Action

    Vendor impersonation: detect payment change requests from vendor lookalike domains

    Source-described action candidate under “Step 4: Configure Detection Policies”; not verified execution.

  19. 25
    Action

    Account compromise: detect sudden changes in employee email behavior

    Source-described action candidate under “Step 4: Configure Detection Policies”; not verified execution.

  20. 26
    Action

    Supply chain BEC: monitor for impersonation of trusted partners

    Source-described action candidate under “Step 4: Configure Detection Policies”; not verified execution.

  21. 27
    Action

    Configure confidence thresholds for auto-block vs. warning banner vs. analyst review

    Source-described action candidate under “Step 4: Configure Detection Policies”; not verified execution.

  22. 28
    Action

    Step 5: Integrate with Response Workflow

    Source-described action candidate under “Step 5: Integrate with Response Workflow”; not verified execution.

  23. 29
    Action

    Auto-quarantine high-confidence BEC detections

    Source-described action candidate under “Step 5: Integrate with Response Workflow”; not verified execution.

  24. 30
    Action

    Add warning banners for moderate-confidence detections

    Source-described action candidate under “Step 5: Integrate with Response Workflow”; not verified execution.

  25. 31
    Action

    Route suspicious emails to SOC analyst queue for review

    Source-described action candidate under “Step 5: Integrate with Response Workflow”; not verified execution.

  26. 32
    Action

    Integrate with SOAR for automated response playbooks

    Source-described action candidate under “Step 5: Integrate with Response Workflow”; not verified execution.

  27. 33
    Action

    Feed BEC verdicts back into training data for model improvement

    Source-described action candidate under “Step 5: Integrate with Response Workflow”; not verified execution.

  28. 34
    Action

    Tools & Resources

    Source-described action candidate under “Tools & Resources”; not verified execution.

  29. 35
    Action

    Validation

    Source-described action candidate under “Validation”; not verified execution.

Potential risk surfaces

Static findings identify points to inspect before installation. They do not establish that a skill is safe, malicious, or effective.

Items warrant review

27 deterministic lexical finding(s) require review across the collected text. Findings can occur in documentation and do not prove execution or exploitability; binary-only files were not scanned.

Network accessObserved
File writesObserved
Credential useNot observed
Note

File modification is described or implemented

23 occurrences

This line contains a possible file-write or redirection operation. Check destination paths and overwrite behavior; the static review did not write any file. This rule matched 23 location(s) in the collected text; 3 representative location(s) are cited. Matching text may be documentation, and the analyzer did not execute it.

Medium

Network access is mentioned

4 occurrences

A URL or network client appears in this source line. Review destinations, data sent, and trust boundaries; static text cannot establish actual network activity. This rule matched 4 location(s) in the collected text; 3 representative location(s) are cited. Matching text may be documentation, and the analyzer did not execute it.

Where context can be spent

Discovery, activation, and resource layers have different meanings. Exact values appear only when the official token-counting method was run with a named model.

Exact token counts unavailable

All token values, including per-file inventory, are null until the pipeline performs an official exact count with an explicit model.

No token measurement was recorded with the required official method. This dossier does not substitute character counts, file sizes, or estimates.

Controlled evaluation

No controlled evaluation

This skill has only the evidence identified above. No task success, token efficiency, latency, or quality advantage is claimed.

Not tested

Use with the source in view

Install command
Inspect the pinned source at https://github.com/mukul975/Anthropic-Cybersecurity-Skills/blob/54a798831d2266a3ca61ce68a7acb80b81160d57/skills/detecting-business-email-compromise-with-ai/SKILL.md and follow the source repository's installation instructions. No installation was performed or verified by this static analyzer. Source license: Apache-2.0.

Prerequisites

  • Static analysis cannot assess runtime behavior, command success, network reachability, or model output quality.
  • No commands, scripts, network requests, model calls, or controlled skill executions were performed.
  • Category, tags, capabilities, workflows, and bilingual editorial summaries are deterministic candidates pending human review, not verified outcomes.
  • Exact token counts are unavailable until a later pipeline performs and records an official model-specific count.

Example prompts

Not provided

A decision, with its limits attached

Strengths

  • The snapshot contains 8 collected file(s), 5 Markdown file(s), and 2 script file(s), enabling source-level inspection. [Evidence: skills/detecting-business-email-compromise-with-ai/SKILL.md:L1]

Considerations

  • No functional, security, or model-quality result has been established. Human review must inspect source context and all cited findings before installation or publication.
Research conclusion

Provisional deterministic static inventory only; pending human editorial review and, separately, controlled evaluation. No install recommendation, security assurance, or effectiveness claim is made.

Traceable by construction

The dossier is a time-bound reading of a public artifact, not an endorsement. Use the source link to inspect the exact upstream material.

mukul975Anthropic-Cybersecurity-Skillsskills/detecting-business-email-compromise-with-ai
Source type
Community
Commit
54a798831d2266a3ca61ce68a7acb80b81160d57
Collected
20 Nov 2026
Last checked
20 Nov 2026
Upstream license
Apache-2.0
Repository stars
36,446
Content basis

Deterministic lexical parsing, YAML frontmatter validation, file inventory, source-described workflow extraction, and heuristic provisional classification of the supplied files only. Generated by machine; no LLM, execution, dynamic testing, or human editorial approval.