docs: add benchmarks directory

[0xallam]

Summary

• Add benchmarks directory with XBEN results overview
• Links to full benchmark repo at usestrix/benchmarks

Details

• 96% success rate on XBOW benchmark (100/104 challenges)
• Strix v0.4.0 in black-box mode
• Performance breakdown by difficulty level
• Resource usage stats

[greptile-apps]

Greptile Summary

This PR adds a new benchmarks directory with documentation of Strix's performance on the XBEN security benchmark suite. The documentation includes a 96% success rate (100/104 challenges) achieved by Strix v0.4.0 in black-box mode, performance breakdown by difficulty level, and resource usage statistics.

Changes:

• New benchmarks/README.md file documenting XBEN benchmark results
• References to the external usestrix/benchmarks repository for full details
• Performance metrics and visualization using a mermaid pie chart
• Links to external benchmark resources on GitHub

Issue found:

• Inconsistent benchmark naming: line 13 refers to "XBOW benchmark" but should be "XBEN" to match the heading, table, and link text.

Confidence Score: 4/5

• This PR is safe to merge with a minor naming inconsistency that should be corrected.
• The PR introduces documentation for benchmark results, which is a low-risk change. The content is clear and informative with proper links to external resources. One syntax issue was found: the benchmark is referred to as "XBOW" in the description but "XBEN" elsewhere, causing an inconsistency. This is a simple fix that should be corrected before merging. The rest of the content including statistics, tables, and formatting is accurate and well-structured.
• benchmarks/README.md - correct the "XBOW" naming to "XBEN" on line 13

Important Files Changed

Filename	Overview
benchmarks/README.md	New benchmarks directory with documentation of Strix's performance on XBEN benchmark. Contains 96% success rate results, performance breakdown by difficulty, and resource usage metrics. One naming inconsistency found: "XBOW" vs "XBEN" on line 13 should be corrected for consistency.

[greptile-apps]

_{1 file reviewed, 1 comment}

_{Edit Code Review Agent Settings | Greptile}

[greptile-apps]

syntax: The benchmark is referred to as "XBOW" here but the heading and links use "XBEN". This inconsistency may confuse readers. Should be consistent - likely should be "XBEN" based on the table and link text above.

Suggested change

	The [XBOW benchmark](https://github.com/usestrix/benchmarks/tree/main/XBEN) is a set of 104 web security challenges designed to evaluate autonomous penetration testing agents. Each challenge follows a CTF format where the agent must discover and exploit vulnerabilities to extract a hidden flag.
	The [XBEN benchmark](https://github.com/usestrix/benchmarks/tree/main/XBEN) is a set of 104 web security challenges designed to evaluate autonomous penetration testing agents. Each challenge follows a CTF format where the agent must discover and exploit vulnerabilities to extract a hidden flag.

Prompt To Fix With AI

This is a comment left during a code review.
Path: benchmarks/README.md
Line: 13:13

Comment:
**syntax:** The benchmark is referred to as "XBOW" here but the heading and links use "XBEN". This inconsistency may confuse readers. Should be consistent - likely should be "XBEN" based on the table and link text above.

```suggestion
The [XBEN benchmark](https://github.com/usestrix/benchmarks/tree/main/XBEN) is a set of 104 web security challenges designed to evaluate autonomous penetration testing agents. Each challenge follows a CTF format where the agent must discover and exploit vulnerabilities to extract a hidden flag.
```

How can I resolve this? If you propose a fix, please make it concise.