analyzing-office365-audit-logs-for-compromise
Parse Office 365 Unified Audit Logs via Microsoft Graph API to detect email forwarding rule creation, inbox delegation, suspicious OAuth app grants, and other indicators of account compromise.
57
48%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/analyzing-office365-audit-logs-for-compromise/SKILL.mdQuality
Discovery
82%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a strong, specific description that clearly identifies the data source (O365 Unified Audit Logs via Graph API) and lists concrete detection capabilities. Its main weakness is the absence of an explicit 'Use when...' clause, which would help Claude know exactly when to select this skill. The domain-specific terminology is well-chosen and would naturally match security analyst queries.
Suggestions
Add an explicit 'Use when...' clause, e.g., 'Use when investigating Office 365 account compromise, analyzing audit logs, reviewing suspicious email rules, or examining OAuth consent grants.'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: 'detect email forwarding rule creation, inbox delegation, suspicious OAuth app grants, and other indicators of account compromise' along with the specific data source 'Office 365 Unified Audit Logs via Microsoft Graph API'. | 3 / 3 |
Completeness | Clearly answers 'what does this do' with specific parsing and detection capabilities, but lacks an explicit 'Use when...' clause or equivalent trigger guidance, which caps this dimension at 2 per the rubric. | 2 / 3 |
Trigger Term Quality | Includes strong natural keywords users would say: 'Office 365', 'Unified Audit Logs', 'Microsoft Graph API', 'email forwarding rule', 'inbox delegation', 'OAuth app grants', 'account compromise'. These are terms a security analyst would naturally use. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with a clear niche: Office 365 audit log analysis for security incident detection via Microsoft Graph API. Very unlikely to conflict with other skills given the specific platform, API, and security-focused use cases. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
14%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill reads as a high-level outline or table of contents rather than an actionable skill. It completely lacks executable code, concrete API endpoints, example queries, or any specific technical detail that would enable Claude to actually perform the described analysis. The workflow is abstract and missing validation steps critical for a multi-step investigation process.
Suggestions
Add executable Python code for each major step: MSAL authentication, Graph API queries with specific endpoints (e.g., `https://graph.microsoft.com/v1.0/users/{id}/mailFolders/inbox/messageRules`), and example filter parameters for suspicious operations.
Include concrete examples of what suspicious audit log entries look like (e.g., example JSON responses) and specific field values to flag (e.g., `ForwardTo`, `RedirectTo` in inbox rules).
Add validation checkpoints: verify authentication succeeded (check token response), confirm audit log query returned results, validate that enumerated rules match expected schema before scoring.
Either flesh out each step with detailed inline guidance or create referenced sub-files (e.g., `GRAPH_AUTH.md`, `AUDIT_QUERIES.md`, `DETECTION_RULES.md`) with clear navigation from the main skill.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The 'When to Use' section is largely filler that Claude doesn't need (e.g., 'When investigating security incidents that require analyzing office365 audit logs for compromise' is circular). The overview also restates what the steps will cover. However, it's not egregiously verbose—just has unnecessary padding. | 2 / 3 |
Actionability | There is zero executable code, no API endpoints, no example queries, no concrete commands. The steps are entirely abstract descriptions ('Query Unified Audit Log for suspicious operations') with no actual Graph API URLs, filter parameters, Python code, or example payloads. This is a description of what to do, not instructions on how to do it. | 1 / 3 |
Workflow Clarity | While steps are numbered, they lack any validation checkpoints, error handling, or feedback loops. For a multi-step investigation involving API calls and data correlation, there's no guidance on what to do if authentication fails, if audit logs are empty, or how to verify results. No concrete sequencing details are provided. | 1 / 3 |
Progressive Disclosure | The content is a flat, monolithic outline with no references to supplementary files, no code examples split out, and no navigation structure. For a skill this complex (7 distinct steps spanning authentication, multiple query types, and report generation), there should be either inline detail or clear references to detailed sub-documents. | 1 / 3 |
Total | 5 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
- Repository
- mukul975/Anthropic-Cybersecurity-Skills
- Commit
888bbe4
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.