Plugin Claude Code
Bounded automatic mixed-model orchestration for orichum.
Plugin Claude Code
Bounded automatic mixed-model orchestration for orichum.
Agent
Read-only high-risk architecture adjudication for security, auth, concurrency, migration, and irreversible design.
Agent
Read-only model-diverse correctness and regression-risk critic.
Agent
Isolated implementation worker for a written plan with an exact disjoint ownership boundary.
Agent
Read-only audited planner for implementation and operational plans with explicit validation, rollback, and stop conditions.
Agent
Read-only allowed bounded replacement for generic repository exploration, with concise file and line evidence.
Agent
Read-only independent verification of a diagnosis, change, or supplied evidence.
Hook
Runs before the agent uses a tool for Agent and Workflow tool calls, executing guard-orchestration.sh (2 commands). From orichum/orichum.
Hook
Runs after the context is compacted for manual and auto tool calls, executing save-compaction-checkpoint.sh. From orichum/orichum.
Hook
Runs when a session starts on startup, resume and compact, executing check-local-services.sh and restore-compaction-checkpoint.sh (2 commands). From orichum/orichum.
Skill Claude CodeCodex
Guidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults.
Skill Claude CodeCodex
Adaptively route independent investigation, review, or high-impact cross-checking to an audited read-only Orichum Workflow when it can add distinct evidence. Do not use when the controller already has sufficient evidence and no independent workflow can add non-duplicative value.