shobman

Audit delivery artifacts against AIDOS rubrics. Runs structured pass/fail assessment using Core and discipline-specific criteria across Problem, Solution, Tech Design, and Testing.

Scaffold an Epic-scope or Feature-scope decomposition into Features and Stories. Reads approved upstream artifacts (Problem + Solution, or feature.md), proposes a decomposition interactively, then commits stubs to the filesystem with TL;DRs and Breakdown Context seeded. Audits the result against the Breakdown rubric. No project-management concerns — audits shape, not sizing or sequencing.

Build delivery artifacts using the AIDOS framework. Scaffolds Problem, Solution, Tech Design, and Testing documents at Epic, Feature, or Story scale with structured rubric-ready output.

Orchestrate sub-agents to fill out per-Feature and per-Story artifacts in an .aidos/ folder whose breakdown has been committed and audited clean. Two-phase dispatch at Epic scope (Features first, then Stories); single-phase at Feature scope (Stories only). Context-isolated, bounded parallelism, autonomous per-artifact build–audit–fix loops.

Review ByteRover project context for generalisable insights and promote them to SAKE collective knowledge. Use at task completion, before compaction, or when asked to share learnings.

Query and contribute to SAKE — a shared knowledge base of operational insights learned by AI agents across sessions. Use before starting tasks to check for known pitfalls, and after discovering non-obvious learnings to share them.