pt-fuzzing-web-api

Performs authorized fuzzing of web applications and APIs to discover input validation failures, parser bugs, and stability issues. Use when testing HTTP endpoints, request parameters, payload handling, and error behavior under malformed or unexpected inputs.

1.41x

Quality

92%

Does it follow best practices?

Impact

99%

1.41x

Average score across 3 eval scenarios

Securityby

Risky

Do not use without reviewing

Evaluation results

100%

21%

Pre-Launch Security Review: Payment Processing API

Fuzzing engagement plan and scope definition

Criteria

Without context

With context

In-scope endpoints defined

100%

Auth contexts specified

100%

Rate/concurrency ceilings

30%

100%

Timeout ceiling specified

80%

100%

Stop conditions defined

100%

Parameter inventory breadth

100%

File upload parameters included

100%

Baseline valid requests

100%

Mutation strategy dimensions

100%

Low-and-slow execution approach

20%

100%

Tracking anomaly types

50%

100%

99%

27%

Security Findings Report: User Profile API Anomalies

Output template compliance and evidence documentation

Criteria

Without context

With context

Scope and Configuration section

60%

100%

Rate/concurrency and stop conditions

100%

Findings section present

100%

Endpoint and trigger input per finding

100%

Observed behavior and repro steps per finding

100%

Security/availability impact per finding

100%

Recommended fix per finding

100%

Stability and Detection Notes section

100%

Regression Test Cases section

10%

100%

Exact trigger payload preserved

71%

92%

True defects separated from expected failures

100%

41%

API Security Triage: Overnight Fuzzing Results

Triage classification and remediation recommendations

Criteria

Without context

With context

Defect vs. validation separation

100%

Minimal payload preserved per defect

100%

Input/schema validation hardening recommended

100%

Parser or library update recommended

100%

Safer defaults recommended

100%

Regression test cases included

100%

Expected safe behavior per regression case

100%

Prevention guidance included

80%

100%

Exact request/response evidence

30%

100%

Security/availability impact stated

100%

Repository: santosomar/ethical-hacking-agent-skills
Commit: a8ff73a

Evaluated: about 1 month ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Pre-Launch Security Review: Payment Processing API Security Findings Report: User Profile API Anomalies API Security Triage: Overnight Fuzzing Results

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.