agile
Deliver a large or multi-stage effort through Hydraia Agile Mode — decompose into an epic, user stories, tasks, subtasks + QA, optionally sync to Jira, then…
Resume an interrupted Hydraia pipeline from the last incomplete phase
> /plugin marketplace add jdanigo/hydraia > /plugin install hydraia@hydraia
How it fires
How this command gets triggered: by you, by Claude, or both.
/resumeContext preview
What this command does when you run it.
Resume an interrupted Hydraia pipeline from the last incomplete phase
description: Resume an interrupted Hydraia pipeline from the last incomplete phase argument-hint: [optional run-log name; defaults to the most recent]
Invoke the **hydraia** skill in resume mode.
0. Resolve the artifacts base first (in-repo `docs/hydraia/` or the external dir the user chose at the storage gate) and inherit that run's **storage location** and **auto-commit** choice from the run log if recorded; if neither is recorded, re-run the Storage & commit gate. Every `docs/hydraia/...` path below is relative to the resolved base. 1. Find the run log: if $ARGUMENTS names one, use `<artifacts base>/runs/$ARGUMENTS`; otherwise use the most recently modified file in `<artifacts base>/runs/`. 2. Read it to recover the original request, the linked plan under `docs/hydraia/plans/`, and the phase checklist. Do NOT re-plan — reuse the frozen plan and spec. Also recover the recorded **run-controls** — execution routing (Balanced/Economy/Max-quality/Hand-off), review depth, summary depth, and E2E strategy — and honor them for the remaining phases. If the run stopped at Phase 3 (a `plan`/`story` route or Hand-off) and no routing/E2E choice was recorded, run the Phase-3 run-controls picker ONCE now — before entering the autonomous half — so execution proceeds without pausing afterward.
2a. **Agile Mode run?** If an epic exists under `<artifacts base>/epics/<slug>/` with a `sprint-status.yaml`, this is an Agile run. Do NOT re-decompose — reuse the frozen tree (`epic.md`, `epic-context.md`, `stories.yaml`, the per-story files). Read `sprint-status.yaml` to find `current` and its `phase`, and continue `phases/agile-orchestrator.md` from that story's first incomplete phase, honoring each story's recorded `tier` and `routing` and the epic's Jira target. Re-sync Jira (`phases/jira-sync.md`) for any story whose local `status` advanced while Jira was unreachable. Tier-L / risky stories still pause for approval on resume. 3. Continue the pipeline from the FIRST unchecked phase, obeying the continuous-execution rule (do not stop until the plan is complete or a genuine blocker appears). 4. Update the run log's checkboxes as phases complete. 5. On completion, print the credits line, exactly: `— Harness By José Daniel Garcés Ospina | Spec Drive Design First —`
An agentic development harness for Claude Code. **One command runs the entire feature pipeline** — it collaborates with you on the design, then builds autonomously: plan, execute, double-review, and verify.
Repo: jdanigo/hydraia
Deliver a large or multi-stage effort through Hydraia Agile Mode — decompose into an epic, user stories, tasks, subtasks + QA, optionally sync to Jira, then…
Design a new app or service from scratch — guided elicitation, architecture proposals, stack choice, API contract, ADRs — then build it through the full…
Launch the local Hydraia dashboard (status, usage telemetry, editable config) in the browser
Diagnose and fix a database bottleneck evidence-first — EXPLAIN, indexes, N+1, locks, pooling, safe migrations
Author CI/CD, containers, or IaC following the repo's platform — deploy and secrets flagged for human approval, never executed
Sync README, API docs, CHANGELOG, and the ADR index with the actual code — reports drift, never invents behavior