Menu
Auto-chained research follow-up skill with recursion guard for residual unknowns.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Model-only playbook for adding a property-based test that pins a function's invariant.
Dispatch a claimed wave to a runtime per the V8 session-reuse ladder.
Fresh-context verification of a phase deliverable or wave outcome. Spawns a fresh auditor subagent that re-reads the diff against the success criteria.
Resolve the Co-Authored-By trailer policy for the active repo.
Compress the session conversation when context approaches the limit.
Read-only design pass for an interactive surface: produces a statechart + action x context matrix + journey scenarios backed by an 11-rule lint contract. No state mutations.
Basé sur la classification professionnelle SOC
| name | blitz |
| description | Auto-chained research follow-up skill with recursion guard for residual unknowns. |
| argument-hint | [--residual-unknowns=<n>] |
| user-invocable | true |
| disable-model-invocation | false |
EAWF_BLITZ_DEPTH_COUNTER) against EAWF_BLITZ_DEPTH (default 8)./research action with blitz=false so the next research pass does not recurse indefinitely./research produced more than one residual unknown.Skill envelope with header.skill = "/blitz". Body carries depth, depth_cap, residual_unknowns, followup_research_args, and next_actions.