diff --git a/src/SUMMARY.md b/src/SUMMARY.md index ddb0a95a157..05478d4e42e 100644 --- a/src/SUMMARY.md +++ b/src/SUMMARY.md @@ -609,6 +609,7 @@ # 🕸️ Pentesting Web +- [Burp Intruder Response Clustering](pentesting-web/burp-intruder-response-clustering.md) - [Web Vulnerabilities Methodology](pentesting-web/web-vulnerabilities-methodology.md) - [Reflecting Techniques - PoCs and Polygloths CheatSheet](pentesting-web/pocs-and-polygloths-cheatsheet/README.md) - [Web Vulns List](pentesting-web/pocs-and-polygloths-cheatsheet/web-vulns-list.md) diff --git a/src/pentesting-web/burp-intruder-response-clustering.md b/src/pentesting-web/burp-intruder-response-clustering.md new file mode 100644 index 00000000000..3983dac11d4 --- /dev/null +++ b/src/pentesting-web/burp-intruder-response-clustering.md @@ -0,0 +1,34 @@ +# Burp Intruder Response Clustering (Colonel Clustered) + +{{#include ../banners/hacktricks-training.md}} + +Colonel Clustered is a Burp Suite extension that **clusters Intruder responses by content similarity** instead of metadata (size/status/type/time), surfacing outliers that otherwise blend in when all responses look the same. + +## Why use it +- Intruder sorting on size/status/time misses **needle-in-the-haystack content changes** (e.g., a single line different across thousands of identical-length responses). +- Extension **tokenizes responses per content-type**, pre-groups to trim work, then clusters the tokenized bodies so outliers appear as tiny clusters (often singletons). +- Parameters are **auto-calibrated per batch** because the whole Intruder result set is available at analysis time—no knobs to tune. + +## Workflow +1. Run an Intruder attack normally. +2. In the Intruder results table, **select all requests → right click → Send to Colonel Clustered**. +3. Open the **Col. Clustered** tab: + - **Fast/Default pass:** automatically starts a **DBSCAN-based** clustering. Shows a progress bar and can be cancelled. + - **Deep Analysis:** click when clustering looks off. Builds a **full similarity matrix** (more accurate, much slower—avoid ~10k+ requests). Also cancellable. +4. Review clusters: choose a cluster to list members, then select a request/response to view it. +5. For a suspected outlier, send it and a “normal” response to **Burp Comparer** to diff the exact body changes. + +## Usage notes +- **Outlier discovery:** identical-size responses that differ in a single body line become a 1-member cluster—trivial to spot, then diff. +- **Accuracy vs. speed:** DBSCAN is usually sufficient; if distinct responses are lumped together, rerun with Deep Analysis to separate them. +- **Scale caution:** the similarity-matrix mode is CPU-heavy; keep large (>~10k) Intruder batches on the fast mode or prune before deep analysis. + +## Install +- Prebuilt JAR: https://github.com/hoodoer/ColonelClustered/releases/download/v1.0.0/ColonelClustered.jar +- Source/README: https://github.com/hoodoer/ColonelClustered + +## References +- [Colonel Clustered: Finding Outliers in Burp Intruder](https://trustedsec.com/blog/colonel-clustered-finding-outliers-in-burp-intruder) +- [Colonel Clustered GitHub](https://github.com/hoodoer/ColonelClustered) + +{{#include ../banners/hacktricks-training.md}} diff --git a/src/pentesting-web/web-vulnerabilities-methodology.md b/src/pentesting-web/web-vulnerabilities-methodology.md index 3bdd35bb45d..886f084e492 100644 --- a/src/pentesting-web/web-vulnerabilities-methodology.md +++ b/src/pentesting-web/web-vulnerabilities-methodology.md @@ -220,6 +220,7 @@ Modern applications extend into browsers, wallets, and automation pipelines—ke - [ ] [**dApps / Decentralized Applications**](dapps-DecentralizedApplications.md) - [ ] [**Browser Extension Pentesting**](browser-extension-pentesting-methodology/) - [ ] [**wfuzz Web Fuzzing**](web-tool-wfuzz.md) +- [ ] [**Burp Intruder Response Clustering (Colonel Clustered)**](burp-intruder-response-clustering.md) ## References