analyzing-malware-persistence-with-autoruns
Use Sysinternals Autoruns to systematically identify and analyze malware persistence mechanisms across registry keys, scheduled tasks, services, drivers, and startup locations on Windows systems.
67
60%
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-malware-persistence-with-autoruns/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 description with excellent specificity and distinctiveness, clearly naming the tool (Sysinternals Autoruns), the task domain (malware persistence analysis), and specific target areas (registry keys, scheduled tasks, services, drivers, startup locations). Its main weakness is the absence of an explicit 'Use when...' clause, which would help Claude know exactly when to select this skill over others.
Suggestions
Add a 'Use when...' clause such as 'Use when the user asks about malware persistence, autoruns analysis, suspicious startup entries, or investigating Windows persistence mechanisms.'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions and locations: 'identify and analyze malware persistence mechanisms across registry keys, scheduled tasks, services, drivers, and startup locations on Windows systems.' | 3 / 3 |
Completeness | Clearly answers 'what' (identify and analyze malware persistence mechanisms using Autoruns across specific locations), but lacks an explicit 'Use when...' clause or equivalent trigger guidance, which caps this at 2 per the rubric. | 2 / 3 |
Trigger Term Quality | Includes strong natural keywords users would say: 'Autoruns', 'malware', 'persistence', 'registry keys', 'scheduled tasks', 'services', 'drivers', 'startup locations', 'Windows'. These cover the domain well and match how analysts would phrase requests. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive due to the specific tool (Sysinternals Autoruns), the specific domain (malware persistence), and the specific platform (Windows). Very unlikely to conflict with other skills. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
37%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill provides a useful automated scanning script but significantly under-delivers on its stated scope. The overview promises baseline diffing, VirusTotal integration, and offline forensic analysis, but the body only contains a single parsing script labeled 'Step 1' with no subsequent steps. The workflow is incomplete with no validation checkpoints or error recovery, and several sections ('When to Use', 'Validation Criteria') contain generic statements rather than actionable guidance.
Suggestions
Complete the workflow by adding Steps 2-4 covering baseline comparison (autorunsc diff), VirusTotal hash checking, and offline analysis with the -z flag—all features promised in the overview.
Add explicit validation checkpoints after each step, such as verifying the CSV output is non-empty, confirming VirusTotal API responses, and a feedback loop for investigating flagged entries.
Remove or condense the generic 'When to Use' section—it adds no value for Claude and wastes tokens on obvious statements.
Add a concrete CLI example for creating and comparing baselines (e.g., `autorunsc64.exe -a * -c > baseline.csv` and the diff workflow), since this is a core advertised capability.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The overview packs useful technical detail efficiently, but the 'When to Use' section is generic filler that Claude doesn't need (e.g., 'When SOC analysts need structured procedures for this analysis type' is vague padding). The prerequisites list is reasonable but could be tighter. | 2 / 3 |
Actionability | The Python script is executable and concrete, which is good. However, the skill only covers automated scanning/parsing—it lacks CLI examples for baseline creation, comparison (the compare/diff feature mentioned in the overview), VirusTotal integration (mentioned in prerequisites but never shown), and offline analysis with -z flag. Key advertised capabilities have no actionable guidance. | 2 / 3 |
Workflow Clarity | Despite being labeled 'Step 1', there is no Step 2 or subsequent steps—the workflow is incomplete. There are no validation checkpoints, no feedback loops for when suspicious entries are found, no guidance on what to do after flagging entries, and no baseline comparison workflow despite it being mentioned in the overview and validation criteria. | 1 / 3 |
Progressive Disclosure | The content has reasonable section structure and external references at the bottom. However, the single large Python script is inline when it could be referenced as a separate file, and there's no navigation to deeper content for the missing workflows (baseline comparison, VirusTotal integration, offline analysis). | 2 / 3 |
Total | 7 / 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
c15f73d
- 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.