adr
Use when a design decision is made that is expensive to reverse, affects multiple components, or rejects a plausible alternative — record it as an Architecture…
Use whenever the user asks to build, add, implement, or change a feature or functionality — or brings a user story or ticket to analyze, reports a bug or unexpected behavior, asks for a branch review, or wants a new app or service designed from scratch. Phase -1 triages the
$ npx -y skills add jdanigo/hydraia --skill hydraia --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/hydraiaContext preview
The summary Claude sees to decide when to auto-load this skill.
Use whenever the user asks to build, add, implement, or change a feature or functionality — or brings a user story or ticket to analyze, reports a bug or unexpected behavior, asks for a branch review, or wants a new app or service designed from scratch. Phase -1 triages the
name: hydraia description: Use whenever the user asks to build, add, implement, or change a feature or functionality — or brings a user story or ticket to analyze, reports a bug or unexpected behavior, asks for a branch review, or wants a new app or service designed from scratch. Phase -1 triages the intent and routes it. Runs the complete non-negotiable development pipeline end to end — deep analysis and planning, sub-agent execution, and a double code-review loop — without asking the user which model, skill, or step to use. This is the default way features get built.
This skill defines the ONLY approved way to build a feature. Run every phase, in order, automatically. **Never ask the user which model, skill, or reviewer to use — those decisions are already made here.** Never skip a phase. Never pause for "should I continue?" between phases.
Announce once at the start: "Running the Hydraia pipeline." Then proceed silently through the phases, narrating at most one short line per phase.
The pipeline body lives in `phases/*.md`, loaded just-in-time. **Read one phase file fully, execute it, then load the next only when its `## NEXT` directs you.**
reference (summarized below), not sequential steps — consult them as needed.
`feature` 0–6 (full) · `plan` 0–3 then stop · `story` -1–3 (PO-first) · `perf`/`db` -1–6 (measurement-first) · `architect` -1–6 (greenfield) · `review` 5–6 · `graph` codegraph only. Ambiguous route → `AskUserQuestion`.
risk overlap, new-logic-vs-mechanical, and UI presence → maps to review depth and ceremony. `autoTier=off` → always Tier M. **Never removes the security floor.**
Quick = trimmed ceremony for a tiny change; Plan = today's full single-goal pipeline; **Agile** = epic-sized / multi-goal / phased work → decompose into Epic→Stories→Tasks→ Subtasks+QA, optional one-way Jira sync, then autonomous stage-by-stage execution gated by autonomy tier (mechanical stories run solo, tier-L/risky pause). Agile branches through `phases/phase-2a-decompose.md` → `phases/agile-orchestrator.md`. Forced by `/hydraia:agile` or `customize.toml` `[agile].mode`.
judgment (triage, design, plan, both reviews). Execution delegates to the Phase-4 executor — **Sonnet 5 by default, overridable via `customize.toml`**. Full policy in `phases/model-policy.md`.
**Balanced** (Sonnet all), **Economy** (Haiku for mechanical + Sonnet for logic/ui), **Max quality** (Opus for logic/ui), or **Hand-off** (freeze the plan, execute later on a cheap external runtime). The picker shows a computed recommendation and what each option costs/gets. Each plan task carries an `Exec class` (mechanical/logic/ui/qa) that Phase 4 maps to a model. `customize.toml` `[executor].routing` can pin a choice and make the run non-interactive.
per-class model / dispatch recipe (read in Phase 4) and the Phase-5 pass-2 reviewer panel (read in Phase 5). Precedence: repo `<artifacts-base>/custom/hydraia.toml` > global `~/.config/hydraia/custom/hydraia.toml` > shipped `skills/hydraia/customize.toml`. Unparseable override → warn and fall back to shipped defaults. **The always-on security gate is NOT customizable.**
`phases/token-discipline.md`): the `caveman` skill compresses internal comms only — never code, specs, or plans.
Read fully and follow `phases/phase--1-triage.md` to begin the pipeline.
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
Use when a design decision is made that is expensive to reverse, affects multiple components, or rejects a plausible alternative — record it as an Architecture…
Teaches the AI to design like a high-end agency. Defines the exact fonts, spacing, shadows, card structures, and animations that make a website feel expensive.…
Use when designing or reviewing an API surface — REST, GraphQL, or gRPC. Contract-first methodology: the contract artifact (OpenAPI 3.1, SDL, proto3) is…
Design banners for social media, ads, website heroes, creative assets, and print. Multiple art direction options with AI-generated visuals. Actions: design,…
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent,…
Brand voice, visual identity, messaging frameworks, asset management, brand consistency. Activate for branded content, tone of voice, marketing assets, brand…