# Malicious AI Agent Skill Detection on 400 labeled skills (89 malicious, 311 benign)

Source: [Malicious AI Agent Skill Detection on 400 labeled skills (89 malicious, 311 benign)](<https://www.sota2.com/research/sota/malicious-ai-agent-skill-detection-on-400-labeled-skills-89-malicious-311-benign>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-09T23:38:02.375Z

Evaluation of model performance for detecting malicious skills using a dataset containing 400 labeled skills (89 malicious, 311 benign). Metrics include Precision, Recall, F1 Score, Accuracy, and False Positive Rate (FPR).

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Malicious AI Agent Skill Detection |
| Dataset (including any stated split) | 400 labeled skills (89 malicious, 311 benign) |
| Metric | Precision |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Precision](<https://www.sota2.com/research/sota/malicious-ai-agent-skill-detection-on-400-labeled-skills-89-malicious-311-benign.md?metric=Precision>): Higher is better; unit: PERCENT.

- [Recall](<https://www.sota2.com/research/sota/malicious-ai-agent-skill-detection-on-400-labeled-skills-89-malicious-311-benign.md?metric=Recall>): Higher is better; unit: PERCENT.

- [F1 Score](<https://www.sota2.com/research/sota/malicious-ai-agent-skill-detection-on-400-labeled-skills-89-malicious-311-benign.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

- [Accuracy](<https://www.sota2.com/research/sota/malicious-ai-agent-skill-detection-on-400-labeled-skills-89-malicious-311-benign.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [False Positive Rate (FPR)](<https://www.sota2.com/research/sota/malicious-ai-agent-skill-detection-on-400-labeled-skills-89-malicious-311-benign.md?metric=False%20Positive%20Rate%20(FPR)>): Lower is better; unit: PERCENT.

## Recorded results

| Method | Precision | Result date | Report paper |
| --- | --- | --- | --- |
| SkillSieve L1 + Single prompt | 100 | Not recorded | 69d7327c7d0172140ec76aba |
| SkillSieve L1 + SSD | 75.2 | Not recorded | 69d7327c7d0172140ec76aba |
| SkillSieve L1 | 58.3 | Not recorded | 69d7327c7d0172140ec76aba |
| ClawVet | 32.9 | Not recorded | 69d7327c7d0172140ec76aba |

### SkillSieve L1 + Single prompt

Precision: 100. Result ID: 69d836b947f34ce924bffb1e.

[Reporting paper](<https://www.sota2.com/research/paper/69d7327c7d0172140ec76aba.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d7327c7d0172140ec76aba.md>)

Recorded evaluation configuration: {"Layer 1":"L1 triage","Layer 2":"Single prompt"}. This record does not assert independent reproduction.

### SkillSieve L1 + SSD

Precision: 75.2. Result ID: 69d836b947f34ce924bffb1f.

[Reporting paper](<https://www.sota2.com/research/paper/69d7327c7d0172140ec76aba.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d7327c7d0172140ec76aba.md>)

Recorded evaluation configuration: {"Layer 1":"L1 triage","Layer 2":"Structured Semantic Decomposition (SSD)"}. This record does not assert independent reproduction.

### SkillSieve L1

Precision: 58.3. Result ID: 69d836b947f34ce924bffb1d.

[Reporting paper](<https://www.sota2.com/research/paper/69d7327c7d0172140ec76aba.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d7327c7d0172140ec76aba.md>)

Recorded evaluation configuration: {"Layer":"L1 triage"}. This record does not assert independent reproduction.

### ClawVet

Precision: 32.9. Result ID: 69d836b947f34ce924bffb1c.

[Reporting paper](<https://www.sota2.com/research/paper/69d7327c7d0172140ec76aba.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
