uuid-ctf
Forensic-CTF UUID recovery tasks solved by an agent in a sandbox. Each task synthesizes a realistic incident filesystem corpus holding five shard UUIDs in five different encodings; the agent must find them, order them by observed_at, derive the final recovery UUID, and write JSON to /workspace/answer.json, which the reward reads and compares against the expected UUID.
Taskset
- Source: Synthetically generated incident filesystem corpus (no dataset) with four difficulty presets
- Size: procedurally generated; count configurable per eval (default 16 samples)
Changelog
- 2026-09-03: Restore default solver network access by reverting the
network_allow=[]default-deny policy introduced in #780; training rollouts need outbound network. - 2026-08-31: Yield task records on demand so bounded evaluations construct only the requested prefix.
- 2026-06-24: Initial v1 taskset.