Skip to content
Apart Research
Apart Research
Contact us
Sprint projectSep 14, 2026San Jose, CA, USA

Watchlines and Muster: Executable Counterfactuals for Cyber Defense

Bobby Faber · Team Bobby Faber with a little help

Submitted to AI Incident Response Sprint. Sprint projects are early-stage work by participants, not Apart Research publications.

Read the report

Report: Watchlines and Muster: Executable Counterfactuals for Cyber Defense

Open in a new tab (opens in new tab)
Code (opens in new tab)
Share

Concepts for reasoning about Cyber Defense: Watchlines

And a modest harness for testing counterfactuals against them: Muster.

Reviews

Judging this Sprint?

Review this project

Your public critique appears on this page without your name. Your private critique is not published; only the Apart team reads it. If you agree below, we share your review with grantmaking.ai (opens in new tab) and the Transformative AI Fund so strong projects can be funded.

Not shown on this page.

Shown on this page, without your name.

Only the Apart team reads this, and funders if you agree below.

Share my name publicly on grantmaking.ai *
Share my private critique with funders *

How much would this matter for AI safety if it worked? How innovative is it? For scores of 4-5: is this actually new to the field, or replicating recent work?

Scoring guide
  1. 1Negligible. No clear problem addressed, or no meaningful novelty.
  2. 2Limited. Addresses a real problem but with a generic or well-trodden approach. Incremental at best.
  3. 3Moderate. Clear problem with a reasonable approach; some novelty in framing or method beyond routine application of existing tools.
  4. 4Significant. Important problem with an original approach, or identifies a neglected problem area. A valuable contribution others could build on.
  5. 5Exceptional. Tackles a critical AI safety problem with a genuinely novel approach, or opens a new research direction. Clear theory of change. You'd be excited to share this with researchers in the area.

How sound are methodology, implementation, and findings?

Scoring guide
  1. 1Seriously flawed. Methodology broken, results uninterpretable, or implementation doesn't work.
  2. 2Weak. Approach has significant gaps: missing validation, flawed experimental design, or incomplete implementation.
  3. 3Competent. Technically solid given the short duration. Methodology makes sense, results are interpretable, limitations acknowledged, work builds toward clear conclusions.
  4. 4Strong. Thorough methodology with convincing validation. Results clearly support conclusions. Immediately useful for future work.
  5. 5Exceptional. Ambitious scope executed rigorously. Surprising findings, novel methods, or unusually robust validation.

How clearly are work, findings, and impact potential communicated?

Scoring guide
  1. 1Incomprehensible. Cannot determine what the project is actually claiming or doing.
  2. 2Hard to follow. Key information buried, missing, or diluted by excessive length. Significant effort to extract main points.
  3. 3Clear enough. Can understand the problem, approach, and results without undue effort. Core content clearly present: problem, method, findings, limitations.
  4. 4Well presented. Easy to follow, well-structured, appropriate level of detail. Target audience would get it quickly.
  5. 5Exceptionally clear. A pleasure to read. Complex ideas made accessible. Could serve as a model for how to present this type of work.

  1. I liked the focus on what remains compromised after a control works. The example where isolation interrupts the evidence needed for later revocation makes that point well.

    It is also good that the report acknowledges when a simple dependency graph gives the same answer. I would align the README with the implemented scenario and test a few alternative attack paths. That would help show how much the conclusions depend on the chosen reconstruction, which the report already recognizes as a limitation.

  2. High clarity tool for testing "if defense is placed here, how does it change attacker capabilities". At this time it only replays attack paths.

    Couple of recommendations/extensions:

    1) Turn into an interactive website for rapid demonstration of the incidents and what controls help with

    2) Potentially enrich the story with failed attempts, denied actions, rejected tool calls, blocked network requests (alternate branches/negative evidence). Informs what is not available.

    3) Leverage same failed attempts (even if synthetic) to also create counterfactual branches, then add plausible alternative branches to see if Watchline still holds

    Nice work!

  3. I think this project makes a useful distinction between stopping an attacker’s next action and removing access they already gained. It replays incident records with proposed defenses to explore how the outcome might change. One example shows how isolating a system early could leave investigators without information needed to remove that access later.

    I would treat this as a useful way to examine response decisions, rather than evidence that early isolation generally makes incidents worse. The result depends on assumptions about when defenses act and what information remains available afterward. Those assumptions need to be clear and consistent, particularly where the paper and repository describe access differently. The example raises a worthwhile question, but broader recommendations need evidence beyond this particular replay.

Cite this project

@misc{faber2026watchlines,
  title = {{Watchlines and Muster: Executable Counterfactuals for Cyber Defense}},
  author = {Bobby Faber},
  year = {2026},
  month = sep,
  note = {Submitted to AI Incident Response Sprint, an Apart Research Sprint},
  howpublished = {\url{https://apartresearch.com/sprints/projects/watchlines-and-muster-executable-counterfactuals-for-cyber-defense-uf3j}},
  url = {https://apartresearch.com/sprints/projects/watchlines-and-muster-executable-counterfactuals-for-cyber-defense-uf3j}
}

More from AI Incident Response Sprint

All 233 entriesBrowse all projects

Build something like this at the next Sprint

AI Collusion Research Sprint · Oct 23 - 25, 2026