Skip to main content
ABLATION

NIST SP 800-218 CISA Secure by Design MIL-HDBK-115C

ABL∀TION is an AI-assisted reverse engineering (AIARE) framework.

Just as ∀ represents a rule that applies to every element without exception.

ABL∀TION stands as the universal constant and definitive answer to reverse engineering by rendering everything before it obsolete.


GUIAI native workbench demo

GUIAI

Ablation's GUIAI unifies GUI with your AI Agent in one native window. The standard Ablation installation includes GUIAI by default; choose the separate CLI-only prompt only if you do not want the workbench.

Install both with the Installation Prompt, or use the platform-specific prompts for Linux, MacOS, and Windows.

Launch with your agent:

guiai desktop /path/to/target --session /path/to/session --agent kimi
# Also: --agent claude, --agent codex, or --agent-command 'your-agent-cli --flag'

Installation Prompt · Platform prompts · CLI only · Workbench guide · Native packages

Installation

Installation Prompt

Copy this prompt into your AI Agent to install the standard Ablation CLI and GUIAI with their dependencies:

Install the standard Ablation CLI and GUIAI, then launch GUIAI with you—the AI Agent I am currently using—embedded in its terminal.

Check my operating system and Python first. Native Windows requires Python 3.11 x64 with Tcl/Tk; macOS and Linux require Python 3.10 or newer with Tkinter. If a prerequisite is missing, explain what is needed and ask before making system changes.

Download and review this installer:
https://github.com/Ablation-Tool/ablation/releases/latest/download/forall.py

Run it with the checked Python. It selects the correct native package, verifies SHA256, and installs the standard Ablation CLI and GUIAI with their required dependencies. GUIAI's wheels install offline; standard Ablation dependencies are downloaded from PyPI. It verifies both ablation --help and guiai --help.

Do not pipe a downloaded script into an interpreter, use sudo, overwrite an existing installation, skip failed tests, or bypass my permissions.

Ask me for the RE target and project/session directory if needed. Keep the project outside the target. On macOS, open the generated Ablation.app. On Windows and macOS, use --toolkit tk when launching through the CLI. Use your actual executable for the embedded AI Agent; if you cannot identify it, ask me instead of substituting another agent.

Do not install an AI model, credentials, optional analysis extras, or firmware analysis. Do not execute the RE target.

Report the operating system, Ablation and GUIAI versions, installation and project paths, validation result, and exact launch command. Confirm that the native window opens and your real CLI starts inside it.

Linux · MacOS · Windows · CLI only

Linux
Install the standard Ablation CLI and GUIAI on Linux, then launch GUIAI with you—the AI Agent I am currently using—embedded in its terminal.

Check for Python 3.10 or newer, Tkinter, a virtual-environment tool, and an X11 or XWayland display. On Ubuntu or Debian, ask before installing missing packages with sudo apt install python3-venv python3-tk python3-gi python3-cairo python3-gi-cairo gir1.2-gtk-3.0 gir1.2-vte-2.91.

Download and review https://github.com/Ablation-Tool/ablation/releases/latest/download/forall.py and run it with the checked Python. Use the default full installation, which includes both Ablation and GUIAI, verifies release checksums, installs GUIAI's wheels offline, and downloads standard Ablation dependencies from PyPI.

Do not overwrite existing files, use sudo for the Python installation, skip failed validation, or execute the RE target. Ask before privileged changes.

Verify ablation --help and guiai --help in the installed environment. Ask me for the RE target and a project directory outside it, then launch with your actual agent executable using guiai desktop TARGET --session PROJECT --agent AGENT_EXECUTABLE. If I need custom agent arguments, use --agent-command instead.

Do not install a model, credentials, or optional analysis extras. Report both versions, installation and project paths, validation results, and the exact launch command. Confirm that the workbench opens and your real CLI starts inside it.
MacOS
Install the standard Ablation CLI and GUIAI on macOS, then open Ablation.app with you—the AI Agent I am currently using—embedded in its terminal.

Check for Python 3.10 or newer with Tkinter. If it is missing, explain how to install a Python distribution that includes Tkinter and ask before making system changes.

Download and review https://github.com/Ablation-Tool/ablation/releases/latest/download/forall.py and run it with the checked Python. Use the default full installation, which includes both Ablation and GUIAI, verifies release checksums, installs GUIAI's wheels offline, and downloads standard Ablation dependencies from PyPI.

Do not overwrite existing files, use sudo for the Python installation, skip failed validation, or execute the RE target. Do not bypass macOS security settings or my permissions.

Verify ablation --help and guiai --help, and run Contents/MacOS/Ablation --self-test from the installation directory. Open the generated Ablation.app. Ask me to choose the RE target, a project directory outside it, and your actual executable. The app is locally built and is not signed or Apple-notarized.

Do not install a model, credentials, or optional analysis extras. Report both versions, installation and project paths, validation results, and the exact app and CLI launch commands. Confirm that the app opens and your real CLI starts inside it.
Windows
Install the standard Ablation CLI and GUIAI on Windows, then launch GUIAI with you—the AI Agent I am currently using—embedded in its terminal.

Check for Python 3.11 x64 with Tcl/Tk support. If it is missing, explain how to install it and ask before making system changes. Do not require WSL.

Download and review https://github.com/Ablation-Tool/ablation/releases/latest/download/forall.py and run it with the checked Python. Use the default full installation, which includes both Ablation and GUIAI, verifies release checksums, installs GUIAI's wheels offline, and downloads standard Ablation dependencies from PyPI.

Do not overwrite existing files, use a global Python installation, skip failed validation, or execute the RE target. Preserve my permissions and existing configuration.

Verify ablation --help and guiai --help in the installed environment. Ask me for the RE target and a project directory outside it, then launch with your actual executable using guiai desktop TARGET --session PROJECT --toolkit tk --agent AGENT_EXECUTABLE. Use --agent-command for custom arguments.

Do not install a model, credentials, or optional analysis extras. Report both versions, installation and project paths, validation results, and the exact launch command. Confirm that the native Windows workbench opens with ConPTY and your real CLI starts inside it.
CLI only

Use this prompt only when you want the Ablation command-line tools without GUIAI. The normal installation includes both by default.

Install only the standard Ablation command-line tools for me, without the GUIAI workbench.

Check for Git and Python 3.10 or newer. Create a project-local virtual environment and activate it for my operating system. For an existing source checkout, verify its origin and preserve uncommitted changes; do not reset, clean, overwrite, or pull over local changes.

Install the standard package with python -m pip install ablation-tool, or from a reviewed source checkout with python -m pip install . Do not run forall.py for this CLI-only installation because its default includes GUIAI.

Keep full, LLM, bytecode, model-provider, credential, and optional analysis extras opt-in. Do not use sudo, global Python installation, alternate package sources, or firmware analysis. Stop on errors.

Verify with ablation --help. Report the checkout and environment paths, Python version, Ablation version, installed extras, and verification result. Do not launch GUIAI or execute the RE target.

Capabilities

Decompilers, Disassemblers & Analyzers

Native CPUs

ISA Variants
Alpha AXP Alpha 21064 · 21164 · 21264 · ECOFF · ELF
ARC ARC EM · ARC HS
ARM ARM-32 · ARM-64 · Thumb
Intel MCS 8051 / MCS-51 · CC254x
LoongArch LoongArch64 · KASAN/KCOV V2
MIPS MIPS-32 · nanoMIPS · MIPS-64
PowerPC PPC-32 · PPC-64 · VLE · PPC476 Book E · XCOFF32 / AIX
RISC-V RV32 · RV64 · HiSilicon ext
S/390 Big-endian S/390 · z/Arch subset · LE/C++ PPA · XPLINK · BFP · Pseudo-C
x86 x86-32 · x86-64
Xilinx MicroBlaze · Versal PDI/CDO · ZynqMP BOOT.BIN

Automotive ECUs

ISA Variants
Infineon TriCore TC1.6x
Infineon C166 C166 · C167
Motorola M68K / CPU32 · MC88100
Renesas SH SH-2A
TI DSP TMS320C28x
V850 / RH850 V850-32 · RH850

VMs & Runtimes

ISA Variants
ARK ArkTS
BEAM Erlang BEAM bytecode
DEX Dalvik · ART · Java-like pseudocode · JNI
Java .class · JAR
MicroPython .mpy v6
Python .pyc (3.8–3.12)
Core Engine
Capability Code Description
AI Knowledge-Graph Analysis 1 2 Combines knowledge graphs with agent-driven retrieval to explore relationships and iteratively refine binary analysis.
Analysis Context & Caching 1 2 Reuses analysis results and session state across runs to avoid redundant processing.
Architecture & ABI Models 1 2 Normalizes x86 instructions and models register aliases, calling conventions, and Linux syscalls in 32-bit and 64-bit modes.
Architecture Definition Language 1 Describes ISA encodings and semantics as reusable graphs, allowing new architectures without changing the core analysis engines.
ARM64 Data-Flow Slicing 1 2 Traces ARM64 store addresses backward to identify exact and possible writes to targeted globals and structure fields.
Binary Parsing & Hardening 1 2 Parses ELF, PE, and Mach-O metadata and inspects ELF hardening, including NX, PIE, RELRO, and stack canaries.
Control & Data-Flow Analysis 1 2 Builds control-flow graphs, propagates dataflow facts, and recovers high-level control structures from branch instructions.
Cryptographic Analysis 1 2 Detects cipher implementations, recovers XOR and affine keys, and tests JWT secrets and RSA/ECDSA nonce reuse.
Debug Metadata & Symbol Recovery 1 2 Recovers ELF function addresses and names from unwind metadata, DWARF, and BTF, plus i386 string cross-references.
Disassembly, Code Profiling & Legacy ARM 1 2 Provides streaming and targeted disassembly, optional Angr-backed dataflow analysis, and legacy ARM PLT recovery.
Dynamic Function Emulation 1 Emulates ARM64 and x86-64 leaf functions to corroborate static findings with observed register and memory writes.
Expression Analysis 1 Converts instructions into typed expressions and merges equivalent computations to reduce redundant taint findings.
Function Discovery 1 2 Recovers function entry points from stripped binaries and flat ROMs using prologues, calls, reset vectors, and alignment patterns.
Function Prioritization 1 2 Profiles function calls and filters findings in functions without runtime call or pointer references.
Intermediate Representation 1 Lifts machine code from supported architectures into a shared IR for dataflow analysis, taint tracking, symbolic execution, and decompilation.
Packed & Encrypted Region Detection 1 Locates compressed, encrypted, and packed regions using sliding-window Shannon entropy analysis.
Register & Argument Tracking 1 2 Tracks symbolic register values across calls and traces sink arguments to network input, file reads, or prior return values.
Static Single Assignment 1 2 Models register definitions with SSA and unifies architecture-specific instructions for type recovery, struct analysis, and pseudo-C generation.
ReAct (Reasoning + Acting)
Capability Code Description
AI-Guided Analysis 1 2 Pre-fetches function context and prior results for single-call LLM naming, role analysis, and patch hypotheses.
Approximate Function Matching 1 2 Indexes opcode-category sequences for fast approximate matching and DTW-based similar-function search.
Cross-Target Learning 1 2 Reuses productive findings, queries, and vendor-specific sink patterns across related targets while discarding unproductive patterns.
Evidence Scoring & Validation 1 2 Ranks probes by expected information gain and executes approved adapters to gather evidence for competing hypotheses.
Function Identification & Naming 1 2 Automatically names known functions using FLIRT-style signatures and preserves analyst-assigned names across binary versions.
Semantic Code Search 1 Ranks functions by natural-language queries using fast vector embeddings, helping analysts locate relevant code before manual disassembly.
Vulnerability Hypothesis Tracking 1 Ranks competing vulnerability hypotheses as evidence accumulates and preserves investigation state across sessions.
Vulnerability Pattern Mining 1 2 Mines prior findings for CWE trends and scans binaries for registered CVE-specific byte patterns.
Security Research
Capability Code Description
Automated Vulnerability Sweeps 1 2 Runs resumable pre-filtering, sink classification, and taint phases, including eight Fortinet vulnerability profiles.
Browser Extension Security 1 Detects externally connectable extensions whose privileged message handlers process web requests without verifying sender identity.
Compiler Defects & Hardening 1 2 Detects LoongArch64 min/max miscompilations and missing or inconsistent binary hardening.
Dangerous Call Classification 1 2 3 Classifies sink arguments on x86-64, ARM32, and LoongArch64 and filters constant-format calls before taint analysis.
Data-Flow Taint Analysis 1 2 3 Traces attacker-controlled data on x86-64, ARM64, and through DEX/JNI boundaries into direct native sinks.
Exploit Development & Scoring 1 2 Builds encoded shellcode and ROP chains, calculates cyclic offsets, and scores findings with BARE for Metasploit modules.
Exploit Path Validation 1 2 Checks path feasibility with bounded Z3 solving and stores reviewed findings for later analysis and export.
Format-String Vulnerabilities 1 Finds non-constant format arguments in logging calls and includes provenance traces for assessing attacker control.
Game Engine Analysis 1 Identifies game engines and engine-specific RCE surfaces in stripped binaries using extensible signature and vtable patterns.
Heap Memory Vulnerabilities 1 2 Detects allocation overflows, use-after-free, double-free, and off-by-one patterns on x86-64 and LoongArch64.
Input Validation & Call Wrappers 1 2 Discovers transitive sink wrappers and profiles sanitization coverage to refine vulnerability triage.
KASAN Finding Validation 1 Verifies KASAN candidates in stripped production binaries, confirming 11 of 14 candidates in TencentOS kernel 6.6.119.
Network Attack-Surface Detection 1 Prioritizes propagated-argument sinks in functions reachable from network input, excluding unreachable candidates from taint analysis.
Parser Length & Bounds Checks 1 2 Detects length-subtraction underflows and validates TLV chunk bounds in protocol parser loops.
Source Code Security Analysis 1 2 Profiles source files and traces dangerous sinks back to request-input sources across repository call graphs.
SQL & Lua Injection Analysis 1 2 Classifies SQL query provenance in binaries and traces request input to Lua evaluation in source trees.
Unauthenticated Attack Surface 1 Maps unauthenticated routes and their reachable sinks by reconstructing dispatch and authentication wrappers from binaries.
Unsafe Input & Function Checks 1 2 Classifies risky sscanf call sites and flags deprecated or unsafe imported functions.
Web & Fortinet Attack Surfaces 1 2 Reconstructs Apache handler chains and maps Fortinet CMDB operations to command-execution sinks.
Windows Driver Vulnerabilities 1 2 Ranks BYOVD primitives by impact and identifies METHOD_NEITHER IOCTL handlers that expose raw user-space pointers.
Firmware Analysis
Capability Code Description
bFLT (Binary Flat Format) 1 2 3 Loads binary Flat executables, recovers ARM32 link-time references, applies Linux RAM relocations, and builds analysis contexts with explicit unresolved outcomes.
CAEN ASCII-Hex Containers 1 Decodes CAEN firmware payloads and verifies their additive checksums without treating integrity checks as signatures.
ECU Architecture Detection 1 2 Identifies raw-ROM architectures and classifies code, calibration data, and interrupt-vector regions before disassembly.
ECU Calibration 1 2 Parses ECU calibration definitions and scaling metadata.
ECU Security Analysis 1 2 Detects predictable UDS and KWP2000 seeds and traces diagnostic input to memory writes on TriCore.
Embedded Architecture Decoding 1 2 Decodes C166/C167 and TI C28x instructions and recovers function starts from ECU firmware.
Extraction & Triage 1 2 Scans firmware signatures, strings, and entropy and extracts partitions from supported vendor containers.
Firmware Steganography 1 Extracts cryptographic material hidden in firmware bitmap pixels.
Fortinet 1 2 Recovers FortiOS partition keys and extracts firmware containers into analyzable ELF and filesystem images.
IBM i / TIMI Save Files 1 2 3 4 Verifies QSYS SAVF records and extracts bounded MI objects; recovers legacy catalogue entries and two non-CL program layouts with explicit, bounded code ranges. TIMI instruction decoding is not yet claimed.
JFFS2 Recovery 1 Reconstructs little-endian JFFS2 files with CRC and extent validation, without mounting or creating device nodes.
Juniper 1 2 3 4 5 6 7 Extracts and verifies signed Junos packages, compares bounded MIPS32/o32 functions, validates VRF gencfg messages, and analyzes ownership and caller/argument provenance.
Legacy Binary Formats 1 2 Parses Motorola M88K COFF binaries and extracts files from Tandem NonStop Guardian PAK archives.
Texas Instruments 1 2 Parses TI COFF2 symbols and detects embedded NNC/TVM models, weak normalization symbols, and NPU usage risks.
Platform & Runtime Analysis
Capability Code Description
Android Cross-Layer 1 2 3 4 5 Provides end-to-end Android analysis: APK/XAPK and DEX decoding, JNI binding recovery, ELF disassembly, and cross-layer taint tracing.
C++ Call Recovery 1 2 Reconstructs C++ class hierarchies and resolves ARM64 and x86-64 virtual calls for accurate call graphs.
Erlang & MicroPython 1 2 Lifts BEAM and MicroPython bytecode to pseudocode and flags dynamic execution and controlled hardware or network operations.
Go Binaries 1 2 Recovers Go function names from pclntab metadata and reconstructs command handlers and message framing in supported binaries.
HarmonyOS Apps 1 2 3 Reconstructs ArkTS modules as JavaScript-like pseudocode and surfaces embedded API keys and endpoint URLs.
IBM AIX 1 2 Parses XCOFF32/64 sections and symbols and extracts BFF package archives without the AIX toolchain.
IBM z/OS 1 2 3 4 5 6 7 Decodes NETDATA and PDS/PDSE members, reconstructs load images, applies RLD relocations, resolves call graphs, emits S/390 pseudo-C, demangles IBM C++ signatures, and compares side-decks and builds without external RE tools.
Java & Spring 1 2 Parses Java artifacts and detects deserialization gadgets, XXE, JNDI injection, and exposed Spring Actuator endpoints.
LoongArch & nanoMIPS 1 2 3 Provides architecture-specific decoding, lifting, taint analysis, and sink classification for LoongArch64 and variable-width nanoMIPS.
PowerPC Call Recovery 1 2 3 Resolves PowerPC indirect calls and ELFv1 trampolines and uses AltiVec density to identify signal-processing routines.
Python, Perl & Electron 1 2 3 Audits Python bytecode, traces Perl input into backticks, and detects insecure Electron windows, IPC, and renderer boundaries.
Windows Binaries 1 2 3 4 5 6 Detects Windows CFG bypasses, exposed SEH handlers, and COM hijacks while extracting ETW providers and RPC interfaces.
Protocol Analysis
Capability Code Description
AI & Agent Security 1 2 Audits AI endpoints, MCP tool flows, RAG stores, and agent-to-agent trust chains for injection and data exposure.
API Security Testing 1 Discovers undocumented APIs and tests major web transports for authorization, injection, SSRF, JWT, and OAuth weaknesses.
CAN Signal Decoding 1 Maps CAN frames to named signals using DBC layouts and scaling metadata.
Nginx Security Testing 1 Fingerprints Nginx and probes for exposed status endpoints, alias traversal, proxy SSRF, and request smuggling.
Protocol State Machines 1 Reconstructs ARM64 protocol state machines.
TLS Security Analysis 1 2 Audits TLS certificates, key exchanges, transport protections, and session-resumption downgrade paths.
Live Ecosystems
Capability Code Description
Cisco Infrastructure Assessment 1 2 Audits Cisco appliance and data-center configurations, recovers weak credentials, and maps trust and lateral-movement paths.
Container & Kubernetes Security 1 2 Assesses container privilege, isolation, secret exposure, RBAC, and cloud-metadata risks.
Forensics & Malware Analysis 1 2 Collects Windows and Linux forensic artifacts, detects timestomping and rootkits, and recovers ELF binaries from memory.
Host & Privilege Assessment 1 2 Inventories hosts, detects process injection, traces syscalls, and identifies privilege-escalation misconfigurations.
Network & Industrial Protocols 1 2 Discovers services and ICS/OT protocols, extracts exposed PCAP credentials, and flags C2 or lateral-movement patterns.
VoIP Security Assessment 1 Enumerates VoIP services and gateways and detects unencrypted media, weak SIP registration, and toll-fraud exposure.
DocIN (Documentation + Integrity)
Capability Code Description
Binary Diffing & Patch Tracking 1 Matches functions across binary versions and reports instruction, callee, and string changes for patch review.
Binary Ninja Plugin 1 Annotates Binary Ninja with cached function names, taint findings, and session verdicts at instruction level.
Finding Export & Reporting 1 2 3 Exports findings as replayable YARA rules, SARIF 2.1.0, and JSON for analysis and reporting workflows.
Module Security Quality Gate 1 Blocks commits with critical or high findings from cached ten-section security audits of Ablation modules.
Sequence-Based Function Matching 1 2 Matches homologous functions with DTW and localizes patched instruction windows using matrix-profile distances.
Software Composition & Supply Chain 1 2 Extracts library versions and provenance into SBOM-ready component records for supply-chain review.
Structural Function Matching 1 Matches functions across architectures and optimized builds by comparing control-flow topology instead of instruction sequences.
Threat Classification 1 Labels findings with MITRE ATT&CK technique and tactic IDs for export and reporting.
Rubric Library
Capability Description
Index A structured collection of named, versionable LLM "behavior specifications" that you can apply to different tasks to get consistent, high-quality analyses on demand.

Setup & About

Acknowledgments

Ablation couldn't exist without these creative individuals.

Though no code was written by these authors, doctors, and engineers, their work has direct influence on the production of code, and the aesthetics that makes Ablation what it is.

Thank you.

-Nick

Research Papers

Title Authors
Reverse Compilation Techniques · Specifying the Semantics of Machine Instructions · UQBT: Adaptable Binary Translation at Low Cost · Machine-Adaptable Dynamic Binary Translation Dr. Cristina Cifuentes
Design of a Retargetable Decompiler for a Static Platform-Independent Malware Analysis Petr Zemek, Lukáš Ďurfina, Jakub Křoustek, Dušan Kolář, Tomas Hruska, Karel Masařík, Alexander Meduna
Finding Taint-Style Vulnerabilities in Linux-based Embedded Firmware with SSE-based Alias Analysis Cheng, Zheng, Liu, Guan, Liu, Li, Zhu, Ye, Sun
iResolveX: Multi-Layered Indirect Call Resolution via Static Reasoning and Learning-Augmented Refinement Monika Santra, Bokai Zhang, Mark Lim, Vishnu Asutosh Dasu, Dongrui Zeng, Gang Tan
Extracting Protocol Format as State Machine via Controlled Static Loop Analysis Qingkai Shi, Xiangzhe Xu, Xiangyu Zhang
NEMETYL: Message Type Identification of Binary Network Protocols using Continuous Segment Similarity Stephan Kleber, Rens Wouter van der Heijden, Frank Kargl
Imperfect Forward Secrecy: How Diffie-Hellman Fails in Practice David Adrian, Karthikeyan Bhargavan, Zakir Durumeric, Pierrick Gaudry, Matthew Green, J. Alex Halderman, Nadia Heninger, Drew Springall, Emmanuel Thomé, Luke Valenta
Nonce-Disrespecting Adversaries: Practical Forgery Attacks on GCM in TLS Hanno Böck, Aaron Zauner, Sean Devlin, Juraj Somorovsky, Philipp Jovanovic
Whitening Sentence Representations for Better Semantics and Faster Retrieval Jianlin Su, Jiarun Cao, Weijie Liu, Yangyiwen Ou
Constant Propagation with Conditional Branches Mark N. Wegman, F. Kenneth Zadeck
A Simple, Fast Dominance Algorithm Cooper, Harvey, Kennedy
libdft: Practical Dynamic Data Flow Tracking for Commodity Systems Vasileios P. Kemerlis, Georgios Portokalidis, Kangkook Jee, Angelos D. Keromytis

Books supplied by www.oreilly.com | github.com/oreillymedia

Title Authors
The Art of Software Security Assessment Mark Dowd, John McDonald, Justin Schuh
Practical Binary Analysis Dennis Andriesse
Practical Malware Analysis Michael Sikorski, Andrew Honig
Practical Reverse Engineering Bruce Dang, Alexandre Gazet, Elias Bachaalany
Hacking: The Art of Exploitation (2e) Jon Erickson
Learning Linux Binary Analysis Ryan O'Neill
Windows Internals Part 1 & 2 Pavel Yosifovich, Mark Russinovich, David Solomon, Alex Ionescu, Andrea Allievi
Rootkits: Subverting the Windows Kernel Greg Hoglund, Jamie Butler
Advanced Compiler Design and Implementation Steven Muchnick
Engineering a Compiler Keith Cooper, Linda Torczon
Practical IoT Hacking Fotios Chantzis, Ioannis Stais, Paulino Calderon, Evangelos Deirmentzoglou, Beau Woods
Inside the Android OS: Building, Customizing, Managing and Operating Android System Services G. Blake Meike
Malware Analysis and Detection Engineering Abhijit Mohanta, Anoop Saldanha
Evasive Malware Kyle Cucci
Hacking Cryptography Kamran Khan, Bill Cox
Real-World Cryptography David Wong
Agentic GraphRAG Anthony Alcaraz, Sam Julien

GUIAI Design and Direction

Dr. Helen R Horkan contributed ideas and design feedback that helped shape GUIAI’s development and the direction of its native workbench. I’m grateful for her insight and for the influence it had on the project.

Honorable Mention

Microsoft Excel (Data Analysis ToolPak). When analyzing closed infrastructure or securing black-box systems, this exact process is called timing analysis or telemetry reverse engineering. Without source code, the Data Analysis ToolPak mathematically deconstructs how an application works on the backend by strictly observing its inputs and outputs.

Real-World Results

Ablation has been used to analyze production firmware and kernel drivers from Cisco, Fortinet, IBM, Juniper, TencentOS, Huawei, and more.

Following coordinated disclosure, the Cisco Product Security Incident Response Team (PSIRT) adopted Ablation for internal vulnerability triage and continues to use it to evaluate ongoing disclosure reports. Cisco Adaptive Security Appliance (ASA) LINA is among the artifacts analyzed, with findings submitted through CERT/CC VINCE.

CVE Product Title CVSS
CVE‑2026‑76420 Secure Firewall Management Center (FMC) Peer Impersonation 9.0 Critical
CVE‑2026‑76412 Secure Firewall Management Center (FMC) Privilege Escalation to root 8.5 High
CVE‑2026‑76413 Secure Firewall Management Center (FMC) Single Sign-On Token Forgery 8.5 High
CVE‑2026‑76447 Identity Services Engine (ISE) OCSP Responder Authentication Bypass 5.3 Medium
Demonstrations

Ablation in Action

Ablation in Action

How to Use Ablation

How to Use Ablation

Learn as You RE

Learn as You RE

Code Standards

NIST SP 800-218

This defines how software should be built, organized around four practice areas: preparing the development environment, protecting the codebase from unauthorized changes, writing software with security built into every layer, and responding when vulnerabilities surface after release.

What this means for Ablation in practice is a mandatory ten-section security audit on every module before it touches the repository. The audit covers input validation, memory handling, integer safety, error handling, concurrency, dependency security, injection prevention, secrets handling, logging, and residual risks. No module is committed without passing it.

Taint trackers, which are the modules that matter most if they get something wrong, carry an additional requirement: 85% statement coverage enforced by a hard gate in the test suite, verified against byte-exact synthetic instruction sequences that test ABI register behavior, not just whether the code runs without crashing. Dependencies are pinned with exact hashes and scanned weekly for known CVEs.

CISA Secure by Design

CISA asks software manufacturers to commit to three things: take ownership of the security outcomes of the people who rely on the software, be transparent about its limitations, and treat security as a requirement rather than a recommendation.

For a RE framework, taking ownership of security outcomes has a specific meaning. A taint tracker that misses a vulnerability does not just produce a wrong answer. It produces a result someone may act on as evidence the binary is clean. That is why a false negative in any Ablation taint tracker is classified as a CRITICAL defect.

Every tracker ships with a documented known-anomalies file describing the exact conditions under which it can produce false negatives, so a researcher knows when to apply manual verification on top of the automated result. The precommit gate is not advisory. No code enters the repository without passing it.

MIL-HDBK-115C

The US Department of Defense handbook for reverse engineering methodology defines four phases:

  • Phase 1: Every engagement starts with a session file recording the target description, artifact hash, tool selection, and documentation plan before any analysis runs. The rule that semantic search runs before manual disassembly maps to the following phase.
  • Phase 2: Requires characterizing a binary at the function level before doing detailed trace work. The DynamicSandbox covers the following.
  • Phase 3: The structured retrospective written at the close of every completed project maps to the final phase.
  • Phase 4: The standards document tracks 29 specific requirements drawn from all three frameworks. All 29 are met.
Responsible Use & Contact

Responsible Use: Ablation is built for authorized security research. Use it only against systems you own or have explicit written permission to test. Running it against systems without authorization violates computer fraud laws in most jurisdictions. The author is not responsible for misuse.

Contact: Feel free to contact me if you have any questions, ideas, complaints, or bugs:

Metadata

Release files for ablation-tool 2.93.2

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for ablation-tool 2.93.2
File Size Uploaded Provenance Actions
ablation_tool-2.93.2.tar.gz 3.2 MB With provenance Details for ablation_tool-2.93.2.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for ablation-tool 2.93.2
File Interpreter ABI Platform Actions
ablation_tool-2.93.2-py3-none-any.whl Python 3 none any Details for ablation_tool-2.93.2-py3-none-any.whl

Total release size: 6.5 MB

Release files / ablation_tool-2.93.2.tar.gz

Download URL ablation_tool-2.93.2.tar.gz
Size 3.2 MB
Tags Source
SHA-256 checksum
How to use checksums
000f88b3d02adf9439e0d3c45fa8479d9b872eaf56e98591776793f35beace72
BLAKE2b-256 checksum
How to use checksums
8de4925e497b654fb07e4c8a265a1a34a768037ba7be3a3454eea6c34d72590a
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 2, 2026.

Transparency log

Release files / ablation_tool-2.93.2-py3-none-any.whl

Download URL ablation_tool-2.93.2-py3-none-any.whl
Size 3.3 MB
Tags Python 3
SHA-256 checksum
How to use checksums
a31f56470c2e4552401bd53292c65e393cb97b7c7cc0220140572cb6a2b8f595
BLAKE2b-256 checksum
How to use checksums
67384ec15d50e350b37b22f4f1c848d56ed1d95d0e25795e0f39a48d86d21409
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 2, 2026.

Transparency log

Security for ablation-tool 2.93.2

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Full provenance

All 2 files in this release have been uploaded with provenance metadata.

Select from 2 files

Attestation for ablation_tool-2.93.2.tar.gz

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 2, 2026.

Transparency log

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page