AI Engineering

Hipocampus Flush dumps the current session context to the daily raw log through a subagent. Use it before compaction when you want to persist what happened in the session.

Hipocampus Recall guides a three-step memory lookup: ROOT.md triage, manifest-based file selection, then qmd search fallback. Use it when a question may depend on past notes or compacted memory.

Hipocampus Search looks up memory with qmd using BM25 and optional vector search, then follows the compaction tree when needed. Use ROOT.md first to decide whether to search memory or look externally.

Pattern Recognition identifies reusable implementation patterns after successful work and records them in the knowledge core. Use it to preserve decisions, capture proven structures, and suggest past patterns before similar tasks.

Planning Methodology creates minimal-change, reversible implementation plans for turning requirements or research into executable blueprints. Use it to map codebase discovery, risks, steps, and verification before implementation.

Plugin Forge scaffolds Claude Code plugins and lints a single plugin or the whole marketplace for manifest, frontmatter, stdlib-only, and test-harness rules. Use it when creating a new plugin or checking existing plugins against house invariants.

Post-Plan Workflow materializes approved plans into MCP items, wires dependencies, and dispatches implementation agents. Use it after plan approval to move from planning into execution and verification.

Quick Start orients users to the MCP Task Orchestrator by detecting workspace state and explaining plan mode, persistent tracking, and project scoping. Use it when you want a first-time walkthrough or need to understand how the pieces work together.

Research Methodology guides version-aware documentation lookup before implementation. It helps when Claude needs authoritative API details, current docs, or migration info instead of relying on stale memory.

Session Retrospective analyzes an implementation run by checking schema quality, delegation alignment, note effectiveness, and plan-to-execution fit. Use it after a run or when you want a session review and improvement notes.

Six-Rung Ladder filters decisions about whether to write code, add a dependency, or introduce an abstraction. It walks through YAGNI, stdlib, platform features, existing dependencies, one-line fixes, and the smallest working version.

Spec Quality Framework defines the minimum analysis a plan must include: alternatives, non-goals, blast radius, risk flags, and test strategy. Use it when filling queue-phase specification notes for MCP work items.