This skill is a specialization of the ctf-sandbox-orchestrator skill for challenges where the hard part is understanding how a firmware image is structured, booted, updated and turned into reachable device behavior. The agent keeps the original image, extracted partitions and patched copies as separate artifacts and records offsets, hashes and entrypoints.
The skill states that it runs only after the orchestrator has set sandbox assumptions.
What You Get
- A three-step workflow: establish image layout, trace boot or update flow, reduce to the decisive path
- A reference checklist for container, partitions, filesystems, bootloader, kernel, init and update verifier stages
- A rule to record offsets and hashes before extraction changes anything
- A hand-off rule to the broader reverse skill when the task shifts to native crash behavior
Common Use Cases
- Unpacking a firmware image from a CTF challenge and mapping its partitions
- Tracing which stage of a boot or update chain consumes a key, certificate or config file
- Documenting how an extracted init script leads to an exposed service in a lab device
What Makes It Unique
It follows the boot or update chain in order and does not jump to the most interesting file. Its checklist warns against editing extracted files before pristine offsets and hashes are recorded, and against calling a secret decisive before showing where the flow consumes it.
Under The Hood
The skill is instruction text only: a SKILL.md, one reference checklist and an agent configuration file. It ships no scripts. The method involves unpacking images, mounting or extracting filesystems and reading config and init files, so the agent’s own tools run commands and write extracted files in the sandbox the orchestrator set up. The skill does not fetch pages or call APIs on its own.