Hook Claude Code
Runs when a tool call fails for TaskUpdate, AskUserQuestion and Elicitation tool calls, executing flow-status-observe.py via python. From Andyduck-ops/omp-flow.
Hook Claude Code
Runs when a tool call fails for TaskUpdate, AskUserQuestion and Elicitation tool calls, executing flow-status-observe.py via python. From Andyduck-ops/omp-flow.
Hook Claude Code
Runs when a subagent finishes for omp-flow-research, omp-flow-architect, omp-flow-qbd, omp-flow-implement and omp-flow-check tool calls, executing flow-status-observe.py via python. From Andyduck-ops/omp-flow.
Hook Claude Code
Runs when a subagent starts for omp-flow-research, omp-flow-architect, omp-flow-qbd, omp-flow-implement and omp-flow-check tool calls, executing inject-agent-identity.py via python (5 commands). From Andyduck-ops/omp-flow.
Settings file Claude Code
Agent settings declaring 6 hook events (SessionStart, PostToolUse, PreToolUse, PostToolUseFailure).
Agent
Converts a selected synthesis into linked design or bounded work Concepts.
Agent
Implements one bounded work Concept and writes its linked handoff.
Agent
Fast read-only scout starting from one Bundle or repository entry.
Agent
Checks a proposed decision against linked Bundle knowledge without reconstructing chat.
Agent
Main coordinator for linked task Bundles and OMP native task operations.
Agent
Produces a bounded approach from one task Bundle entry Concept.
Agent
Independently challenges linked design or work-map Concepts.
Agent
Investigates one bounded question from a task Bundle and writes linked evidence.
Agent
Independently reviews a work Concept, its handoff, and changed code.
Instructions file CodexOpenCode
Instructions for Andyduck-ops/omp-flow, covering agents.md — omp-flow engineering guide, ownership boundary, workflow, bundle contract and runtime and assignment contract.