National Cyber Warfare Foundation (NCWF)login

ashirt-server for centralized red team evidence capture and reporting


0 user ratings
2026-11-14 17:28:24
milo
Red Team (CNA)
"ashirt-server

ashirt-server is the backend and web frontend of ASHIRT, a toolkit that captures, indexes, and searches high-fidelity operation evidence for authorized red teams and adversary simulations.








Toolashirt-ops/ashirt-server — server and frontend for the ASHIRT evidence capture, indexing, and reporting toolkit
CategoryOperation evidence management and reporting platform
Primary UseCentralizing screenshots, notes, and other evidence from multiple operators into a searchable, indexed repository during authorized engagements
Safe UseDocumentation support for authorized red team engagements, adversary simulations, and penetration tests under written engagement scope
Telemetry NoteThe server itself is benign infrastructure; defenders reviewing red team artifacts will find consolidated evidence records rather than new network telemetry

ashirt-server is the server-side and frontend half of ASHIRT, the Adversary Simulators High-Fidelity Intelligence and Reporting Toolkit, written in Go and released under the MIT license by maintainers from Yahoo's offensive security team, John Kennedy and Joe Rozner. The repository currently sits at 171 stars and carries topics like red-team, offensive-security, evidence, and reporting, which accurately describes its niche: it is not an exploitation framework but an evidence lifecycle platform. The companion pieces live in separate repositories — the ASHIRT client at ashirt-ops/ashirt and the terminal integration aterm at ashirt-ops/aterm — with this repo holding only the central server and web UI.


The problem statement in the README will feel painfully familiar to anyone who has closed out a long engagement. Documentation and reporting are essential but universally disliked, and by the time operators get to them, the exact sequence of events is fuzzy and the evidence is scattered. Teams typically patch over this with ad hoc note-taking apps, shared screenshot folders, and manual copying between machines. Those approaches rarely scale, are awkward to share, and break down quickly once evidence spans multiple operators and multiple computers. The motivating failure mode is concrete: digging through a pile of post-operation screenshots trying to find the one image you need, assuming it was captured at all.


ASHIRT's answer is a centralized synchronization point that captures high-fidelity data from all evidence sources during an operation, indexes it, and provides search over the result. Two design adjectives in the README carry most of the architectural weight. The system is described as non-intrusive, meaning it sits alongside the operator's workflow rather than dictarding tooling choices, and automatic when possible, meaning evidence capture hooks into the operator's environment so that recording does not depend on someone remembering to hit a save button. The server is the aggregation and query layer that makes evidence from many sources look like one coherent, searchable corpus.


From a deployment standpoint, the project offers two distinct paths in its Install section. For a local evaluation there is a quickstart document in the repository wiki, which lowers the barrier for a team evaluating the tool before committing to infrastructure. For production, the project points at ashirt-deployments, specifically an ECS reference under that repository's main branch, indicating that the maintainers operate ASHIRT on AWS container infrastructure and are sharing a deployment reference rather than just build instructions. That split between wiki quickstart and production deployment reference suggests the tool is actively used in real engagements, not just published as a curiosity.


The build itself is split across two sub-READMEs: frontend/Readme.md for the web frontend and backend/Readme.md for the Go backend. The backend README also serves as the configuration reference, covering the various components, their configuration options, and how the components interact. Reading the repository layout, you can infer a conventional three-part architecture: a Go API server, a JavaScript frontend served to reviewers and operators, and a client ecosystem — the ASHIRT client and aterm — that pushes evidence in. The server is the contract holder; clients are pluggable producers.


What makes this interesting architecturally is the multi-operator dimension. The README frames evidence as spanning multiple operators and computers, which means the server must handle concurrent ingestion, associate evidence with the right operation and operator, and present a unified timeline or searchable index afterward. That is a genuinely different problem from single-user note-taking tools like CherryTree or Obsidian, which is why ad hoc solutions fail at team scale. The search-over-index capability is the payoff: instead of remembering where a screenshot went, you query the central store for the artifact that proves a specific action.


The team also published a 2020 presentation on ASHIRT, linked from the README, which is worth watching if you are evaluating the tool — it is the only place outside the code where the design rationale is spelled out in depth. The presence of a public talk, named maintainers with corporate email addresses at Yahoo, a Contributing.md that explicitly welcomes issues, questions, and pull requests, and a structured LICENSE setup all point to a maintained, professionally backed project rather than a personal script dump.


In an authorized workflow, ashirt-server sits at the documentation layer, which is arguably the highest-value, lowest-glamour part of a red team engagement. A typical fit is a multi-operator engagement where each operator runs the ASHIRT client or aterm to capture evidence locally, which synchronizes to the central server; the lead then queries the aggregated evidence when assembling the final report. Because the tool captures and stores potentially sensitive operation data, teams deploying it should treat the server as engagement-sensitive infrastructure — the ECS deployment reference implies network isolation and access control decisions that operators will want to get right before pointing clients at it.


For defenders and blue teams, this tool is worth understanding from the opposite direction. Evidence captured in ASHIRT is exactly the material a purple team exercise or engagement debrief would draw from, and its existence is a marker of a mature, documentation-heavy offensive practice. Nothing about ashirt-server generates network attack traffic itself; it is bookkeeping infrastructure. Detection conversations around it are therefore about governance — where such servers live, who can read them — rather than about signatures.


There are honest gaps in what the README alone tells you. There is no feature list for the web UI, no detail on authentication mechanisms, no discussion of evidence retention or export formats, and no screenshots. To answer those questions you would need to read backend/Readme.md, the wiki quickstart, and the deployment reference, or simply stand the stack up locally via the quickstart. What the README does make unambiguous is the scope and intent: a non-intrusive, automatically-fed, centrally indexed evidence store for adversary simulation teams.


For teams that keep losing evidence across operators, machines, and weeks of engagement time, ashirt-server attacks the root cause — manual, decentralized evidence handling — rather than the symptom. It is a Go server with a web frontend, a documented deployment path from local trial to ECS production, a client ecosystem, and professional stewardship. As a documentation and reporting platform for authorized offensive work, it occupies a niche that few other open-source tools address this directly, and the 171-star count suggests a focused but real community finds it valuable.



Official project repository for ashirt-ops/ashirt-server.

Download Tool

Educational analysis for authorized security professionals. Use only in controlled, authorized environments.






Source: OffensiveSec
Source Link: https://www.offsecblog.com/2026/11/ashirt-server-for-centralized-red-team.html


Comments
new comment
Nobody has commented yet. Will you be the first?
 
Forum
Red Team (CNA)



Copyright 2012 through 2026 - National Cyber Warfare Foundation - All rights reserved worldwide.