Back to catalog
Static reviewDevelopment

Skill dossier

implementing-data-loss-prevention-with-microsoft-purview

Source declaration

Implements DLP policies using Microsoft Purview PowerShell cmdlets and the Graph API to protect data across Exchange Online, SharePoint, OneDrive, Teams, endpoints, and Power BI, including sensitivity labels, custom sensitive information types with regex, endpoint DLP rules, and Activity Explorer monitoring. Use when stopping PII/PHI/PCI exfiltration, configuring sensitivity labels, or investigating DLP incidents for policy tuning.

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

Implements DLP policies using Microsoft Purview PowerShell cmdlets and the Graph API to protect data across Exchange Online, SharePoint, OneDrive, Teams, endpoints, and Power BI, including sensitivity labels, custom sensitive information types with regex, endpoint DLP rules, and Activity Explorer monitoring. Use when stopping PII/PHI/PCI exfiltration, configuring sensitivity labels, or investigating DLP incidents for policy tuning.

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.

Implements DLP policies using Microsoft Purview PowerShell cmdlets and the Graph API to protect data across Exchange Online, SharePoint, OneDrive, Teams, endpoints, and Power BI, including sensitivity labels, custom sensitive information types with regex, endpoint DLP rules, and Activity Explorer monitoring. Use when stopping PII/PHI/PCI exfiltration, configuring sensitivity labels, or investigating DLP incidents for policy tuning. 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/implementing-data-loss-prevention-with-microsoft-purview/SKILL.md:L56]
  • Step 1: Design the Sensitivity Label Taxonomy [Source statement; provisional, not tested. Evidence: skills/implementing-data-loss-prevention-with-microsoft-purview/SKILL.md:L58]
  • Step 2: Create DLP Policies with Sensitive Information Types [Source statement; provisional, not tested. Evidence: skills/implementing-data-loss-prevention-with-microsoft-purview/SKILL.md:L124]
  • Step 3: Deploy Endpoint DLP Rules [Source statement; provisional, not tested. Evidence: skills/implementing-data-loss-prevention-with-microsoft-purview/SKILL.md:L237]
  • Step 4: Configure Auto-Labeling Policies [Source statement; provisional, not tested. Evidence: skills/implementing-data-loss-prevention-with-microsoft-purview/SKILL.md:L343]
  • Step 5: Monitor with Activity Explorer and Manage DLP Alerts [Source statement; provisional, not tested. Evidence: skills/implementing-data-loss-prevention-with-microsoft-purview/SKILL.md:L394]

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/implementing-data-loss-prevention-with-microsoft-purview/SKILL.md:L1-L34]

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
4
Markdown files
2
scripts
1
references
1
Size
61.4 KB
Structural complexity
Low

File inventory

PathTypeBytes
skills/implementing-data-loss-prevention-with-microsoft-purview/LICENSEOther11 KB
skills/implementing-data-loss-prevention-with-microsoft-purview/references/api-reference.mdReference5.2 KB
skills/implementing-data-loss-prevention-with-microsoft-purview/scripts/agent.pyScript14.5 KB
skills/implementing-data-loss-prevention-with-microsoft-purview/SKILL.mdInstruction30.7 KB

SKILL.md frontmatter

{
  "name": "implementing-data-loss-prevention-with-microsoft-purview",
  "description": "Implements DLP policies using Microsoft Purview PowerShell cmdlets and the Graph API to protect data across Exchange Online, SharePoint, OneDrive, Teams, endpoints, and Power BI, including sensitivity labels, custom sensitive information types with regex, endpoint DLP rules, and Activity Explorer monitoring. Use when stopping PII/PHI/PCI exfiltration, configuring sensitivity labels, or investigating DLP incidents for policy tuning.\n",
  "domain": "cybersecurity",
  "subdomain": "data-protection",
  "tags": [
    "DLP",
    "Microsoft-Purview",
    "sensitivity-labels",
    "endpoint-DLP",
    "data-classification",
    "compliance"
  ],
  "version": "1.0.0",
  "author": "mukul975",
  "license": "Apache-2.0",
  "nist_csf": [
    "PR.DS-01",
    "PR.DS-02",
    "PR.DS-10",
    "GV.PO-01"
  ],
  "mitre_attack": [
    "T1486",
    "T1530",
    "T1537",
    "T1048",
    "T1573"
  ]
}

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
    Decision

    When to Use

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

  2. 02
    Action

    Workflow

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

  3. 03
    Action

    Step 1: Design the Sensitivity Label Taxonomy

    Source-described action candidate under “Step 1: Design the Sensitivity Label Taxonomy”; not verified execution.

  4. 04
    Action

    Establish label tiers: Create a label hierarchy reflecting data sensitivity levels. A standard enterprise taxonomy includes:

    Source-described action candidate under “Step 1: Design the Sensitivity Label Taxonomy”; not verified execution.

  5. 05
    Action

    Define protection settings per label: For each label, configure encryption scope (all employees, specific groups, or custom permissions), content marking (headers, footers, watermarks), and auto-labeling conditions:

    Source-described action candidate under “Step 1: Design the Sensitivity Label Taxonomy”; not verified execution.

  6. 06
    Action

    Configure content marking: Apply visual indicators that persist with the document:

    Source-described action candidate under “Step 1: Design the Sensitivity Label Taxonomy”; not verified execution.

Inspect all extracted workflow candidates (49 candidate steps)
  1. 07
    Decision

    Publish labels via label policy: Labels must be published to users through a label policy that defines which users see the labels and whether a default label or mandatory labeling is enforced:

    Source-described decision candidate under “Step 1: Design the Sensitivity Label Taxonomy”; not verified execution.

  2. 08
    Action

    Step 2: Create DLP Policies with Sensitive Information Types

    Source-described action candidate under “Step 2: Create DLP Policies with Sensitive Information Types”; not verified execution.

  3. 09
    Action

    Create a DLP policy using built-in sensitive information types: Microsoft Purview includes 300+ built-in SITs for credit card numbers, Social Security numbers, passport numbers, and health records. Create a policy targeting financial data:

    Source-described action candidate under “Step 2: Create DLP Policies with Sensitive Information Types”; not verified execution.

  4. 10
    Action

    Create custom sensitive information types with regex: Define organization-specific patterns for data that built-in SITs do not cover:

    Source-described action candidate under “Step 2: Create DLP Policies with Sensitive Information Types”; not verified execution.

  5. 11
    Action

    Use sensitivity labels as DLP conditions: Create policies that apply different restrictions based on the label applied to the content:

    Source-described action candidate under “Step 2: Create DLP Policies with Sensitive Information Types”; not verified execution.

  6. 12
    Action

    Step 3: Deploy Endpoint DLP Rules

    Source-described action candidate under “Step 3: Deploy Endpoint DLP Rules”; not verified execution.

  7. 13
    Action

    Verify device onboarding: Confirm devices are onboarded to Microsoft Purview endpoint DLP through Microsoft Intune or the local onboarding script:

    Source-described action candidate under “Step 3: Deploy Endpoint DLP Rules”; not verified execution.

  8. 14
    Action

    Configure endpoint DLP settings: Define global settings that control which applications and file types endpoint DLP monitors:

    Source-described action candidate under “Step 3: Deploy Endpoint DLP Rules”; not verified execution.

  9. 15
    Input

    Create endpoint-specific DLP rules: Define rules that control copy-to-USB, print, upload, and clipboard operations for sensitive content:

    Source-described input candidate under “Step 3: Deploy Endpoint DLP Rules”; not verified execution.

  10. 16
    Action

    Configure printer groups and USB device exceptions: Allow specific printers and approved USB devices while blocking unauthorized removable media:

    Source-described action candidate under “Step 3: Deploy Endpoint DLP Rules”; not verified execution.

  11. 17
    Action

    Step 4: Configure Auto-Labeling Policies

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

  12. 18
    Action

    Create auto-labeling policy for email: Automatically label inbound and outbound emails containing sensitive information:

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

  13. 19
    Action

    Create auto-labeling policy for SharePoint and OneDrive: Label existing files at rest that match sensitive information patterns:

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

  14. 20
    Action

    Simulate before enforcing: Always run auto-labeling in simulation mode first. Review the simulation results in the Microsoft Purview portal under Information Protection > Auto-labeling. The simulation shows estimated matches per location and sample content matches for validation. Only switch to enforcement mode after confirming accuracy:

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

  15. 21
    Action

    Step 5: Monitor with Activity Explorer and Manage DLP Alerts

    Source-described action candidate under “Step 5: Monitor with Activity Explorer and Manage DLP Alerts”; not verified execution.

  16. 22
    Action

    Use Activity Explorer and the DLP alerts dashboard to monitor policy effectiveness and investigate incidents:

    Source-described action candidate under “Step 5: Monitor with Activity Explorer and Manage DLP Alerts”; not verified execution.

  17. 23
    Action

    Access Activity Explorer: Navigate to Microsoft Purview portal > Data Classification > Activity Explorer. Filter by activity type "DLPRuleMatch" to see all DLP policy matches. Key columns include:

    Source-described action candidate under “Step 5: Monitor with Activity Explorer and Manage DLP Alerts”; not verified execution.

  18. 24
    Action

    Activity timestamp and user principal name

    Source-described action candidate under “Step 5: Monitor with Activity Explorer and Manage DLP Alerts”; not verified execution.

  19. 25
    Action

    Sensitive information type matched and confidence level

    Source-described action candidate under “Step 5: Monitor with Activity Explorer and Manage DLP Alerts”; not verified execution.

  20. 26
    Action

    Policy and rule name that triggered

    Source-described action candidate under “Step 5: Monitor with Activity Explorer and Manage DLP Alerts”; not verified execution.

  21. 27
    Action

    Action taken (Audit, Block, Warn with Override)

    Source-described action candidate under “Step 5: Monitor with Activity Explorer and Manage DLP Alerts”; not verified execution.

  22. 28
    Action

    Location (Exchange, SharePoint, OneDrive, Endpoint)

    Source-described action candidate under “Step 5: Monitor with Activity Explorer and Manage DLP Alerts”; not verified execution.

  23. 29
    Action

    File name and site URL

    Source-described action candidate under “Step 5: Monitor with Activity Explorer and Manage DLP Alerts”; not verified execution.

  24. 30
    Output

    Analyze false positive rates: Export Activity Explorer data filtered by "Override" actions with justification text to identify rules that users frequently override. A high override rate (>20%) indicates the rule may be too aggressive or matching non-sensitive content:

    Source-described output candidate under “Step 5: Monitor with Activity Explorer and Manage DLP Alerts”; not verified execution.

  25. 31
    Input

    Configure DLP alerts: Set up alert policies in Microsoft Purview > Data Loss Prevention > Alerts to receive notifications for high-severity matches:

    Source-described input candidate under “Step 5: Monitor with Activity Explorer and Manage DLP Alerts”; not verified execution.

  26. 32
    Action

    Query DLP events via Microsoft Graph API: Programmatically retrieve DLP alerts and policy match details for integration with SIEM or custom dashboards:

    Source-described action candidate under “Step 5: Monitor with Activity Explorer and Manage DLP Alerts”; not verified execution.

  27. 33
    Action

    Retrieve DLP policy match details for compliance reporting: Use the unified audit log to extract granular DLP match data including the matched content, SIT type, and confidence level:

    Source-described action candidate under “Step 5: Monitor with Activity Explorer and Manage DLP Alerts”; not verified execution.

  28. 34
    Action

    Key Concepts

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

  29. 35
    Action

    Tools & Systems

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

  30. 36
    Action

    Common Scenarios

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

  31. 37
    Action

    Design a four-tier sensitivity label taxonomy: Public, Internal, Confidential (with sub-labels for Finance, Legal, HR), and Highly Confidential. Publish labels to all users with "Internal" as the default label and mandatory labeling enabled for email.

    Source-described action candidate under “Scenario: Protecting Financial Data Across a Multinational Organization”; not verified execution.

  32. 38
    Output

    Create a DLP policy "PCI-DSS Financial Data Protection" scoped to all Exchange, SharePoint, OneDrive, Teams, and Endpoint locations. Configure two rules: a warning rule for 1-4 credit card numbers (notify user, allow override with justification) and a blocking rule for 5+ credit card numbers (block external sharing, generate incident report to compliance team).

    Source-described output candidate under “Scenario: Protecting Financial Data Across a Multinational Organization”; not verified execution.

  33. 39
    Input

    Deploy endpoint DLP with rules blocking copy-to-USB and upload-to-unapproved-cloud for any file containing credit card numbers or labeled "Confidential - Finance". Allow printing with audit logging. Configure approved USB device exceptions for encrypted corporate drives by vendor/product ID.

    Source-described input candidate under “Scenario: Protecting Financial Data Across a Multinational Organization”; not verified execution.

  34. 40
    Action

    Create auto-labeling policies that scan existing SharePoint finance sites and OneDrive locations, automatically applying "Confidential - Finance" to documents matching credit card number or bank account number SITs with confidence level 85+.

    Source-described action candidate under “Scenario: Protecting Financial Data Across a Multinational Organization”; not verified execution.

  35. 41
    Action

    Run all policies in simulation mode for 14 days. Review Activity Explorer for false positive rates, override patterns, and unprotected sensitive content locations. Tune SIT confidence thresholds from 75 to 85 on the credit card SIT after identifying false positives from partial number sequences in meeting notes.

    Source-described action candidate under “Scenario: Protecting Financial Data Across a Multinational Organization”; not verified execution.

  36. 42
    Action

    Switch to enforcement mode after stakeholder sign-off. Configure DLP alerts with Microsoft Sentinel integration for real-time incident correlation. Schedule monthly Activity Explorer reviews to track policy effectiveness metrics.

    Source-described action candidate under “Scenario: Protecting Financial Data Across a Multinational Organization”; not verified execution.

  37. 43
    Action

    Create a custom sensitive information type using regex `RX-20[2-3][0-9]-[0-9]{4}` with corroborating keywords ("compound", "trial", "formulation", "assay", "efficacy") within 300-character proximity. Set primary pattern at 85% confidence and keyword-corroborated pattern at 95%.

    Source-described action candidate under “Scenario: Implementing Custom DLP for Intellectual Property Protection”; not verified execution.

  38. 44
    Action

    Create a second custom SIT for compound identifiers using regex `CPD-[A-Z]{3}-[0-9]{5}` with keywords ("molecule", "synthesis", "pharmacokinetics") for higher confidence matching.

    Source-described action candidate under “Scenario: Implementing Custom DLP for Intellectual Property Protection”; not verified execution.

  39. 45
    Action

    Deploy a DLP policy scoped to the Research department's Exchange distribution list, SharePoint research site collection, and research team OneDrive accounts. Block external sharing, block forwarding to non-research internal users, and generate alerts for the research compliance officer.

    Source-described action candidate under “Scenario: Implementing Custom DLP for Intellectual Property Protection”; not verified execution.

  40. 46
    Action

    Configure endpoint DLP to prevent copy-to-USB and screen capture of documents containing the custom SITs on research department laptops. Allow printing only to approved secure printers in the research facility.

    Source-described action candidate under “Scenario: Implementing Custom DLP for Intellectual Property Protection”; not verified execution.

  41. 47
    Action

    Create a sensitivity label "Highly Confidential - Research" with encryption restricting access to the Research security group. Configure auto-labeling to apply this label to documents matching either custom SIT.

    Source-described action candidate under “Scenario: Implementing Custom DLP for Intellectual Property Protection”; not verified execution.

  42. 48
    Action

    Monitor Activity Explorer weekly for 30 days post-deployment. The compliance team identifies that the RX-YYYY-NNNN regex matches historical project codes in archived documents. Refine the regex to `RX-202[4-6]-[0-9]{4}` to target only active project codes and reduce false positives by 60%.

    Source-described action candidate under “Scenario: Implementing Custom DLP for Intellectual Property Protection”; not verified execution.

  43. 49
    Output

    Output Format

    Source-described output candidate under “Output Format”; 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.

Attention recommended

53 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 useObserved
Medium

Network access is mentioned

18 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 18 location(s) in the collected text; 3 representative location(s) are cited. Matching text may be documentation, and the analyzer did not execute it.

Note

File modification is described or implemented

31 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 31 location(s) in the collected text; 3 representative location(s) are cited. Matching text may be documentation, and the analyzer did not execute it.

High

Credential or secret material is referenced

4 occurrences

This line resembles a credential literal or reads a sensitive environment value. Do not expose its contents; verify whether it is a placeholder, required secret, or accidentally committed material. 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/implementing-data-loss-prevention-with-microsoft-purview/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 4 collected file(s), 2 Markdown file(s), and 1 script file(s), enabling source-level inspection. [Evidence: skills/implementing-data-loss-prevention-with-microsoft-purview/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/implementing-data-loss-prevention-with-microsoft-purview
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.