Agent
Observability specialist for the Hydraia pipeline. Instruments structured logging, metrics (RED/USE), traces (OpenTelemetry), health checks, and alert rules as plan tasks. Vendor-agnostic — OTel first, vendor SDKs only if already present. Never logs secrets or PII. Opt-in via /hydraia:observability.
Agent
Analyze type design for encapsulation, invariant expression, usefulness, and enforcement.
Agent
Expert TypeScript/JavaScript code reviewer specializing in type safety, async correctness, Node/web security, and idiomatic patterns. Use for all TypeScript and JavaScript code changes. MUST BE USED for TypeScript/JavaScript projects.
Agent
Expert Vue.js code reviewer specializing in Composition API correctness, reactivity pitfalls, component architecture, template security, and Vue-specific performance. Use for any change touching .vue, .ts/.js files with Vue imports, or Vue ecosystem code (Pinia, Vue Router, Nuxt). MUST BE USED for Vue projects.
Skill Claude CodeCodex
Codex-native hydraia pipeline. Use when the user asks to build, add, implement, or change a feature, brings a story/ticket, reports a bug, or wants a branch review. Runs the full spec-drive pipeline on Codex — triage, analyze, spec + threat model, plan + self-review, sub-agent execution, double review, verify — using…
Command
Design a new app or service from scratch — guided elicitation, architecture proposals, stack choice, API contract, ADRs — then build it through the full pipeline.
Command
Launch the local Hydraia dashboard (status, usage telemetry, editable config) in the browser.
Command
Diagnose and fix a database bottleneck evidence-first — EXPLAIN, indexes, N+1, locks, pooling, safe migrations.
Command
Author CI/CD, containers, or IaC following the repo's platform — deploy and secrets flagged for human approval, never executed.
Command
Sync README, API docs, CHANGELOG, and the ADR index with the actual code — reports drift, never invents behavior.
Command
Validate, install, and update Hydraia's external dependencies (codegraph, markitdown, E2E browser binaries).
Command
Generate and run an end-to-end suite for the critical user flows — Playwright, auto-detected, never imposed.
Command
Build a feature through the full Hydraia pipeline (analyze → plan → execute → double-review).
Command
Query the code graph for structure, call sites, and blast radius.
Command
Instrument logging, metrics, traces, health checks, and alerts — OpenTelemetry-first, never logging secrets or PII.
Command
Diagnose and fix a performance problem measurement-first — baseline, profile, numeric target, implement, re-measure.
Command
Run only Hydraia analysis + spec + detailed plan (with self-review loop), then stop.
Command
Resume an interrupted Hydraia pipeline from the last incomplete phase.
Command
Run only the Hydraia double code-review + security gate on the current branch.
Command
Analyze a user story like a Product Owner, then produce spec + QA cases + a frozen plan — no code.
Hook
Runs when a session starts on startup, clear and compact, executing preflight.sh. From jdanigo/hydraia.
Hook
Runs before the agent uses a tool for Edit, Write, MultiEdit, Task and Bash tool calls, executing gate.sh, blastgate.sh, agents.sh and plancheck.sh (4 commands). From jdanigo/hydraia.
Hook
Runs when a subagent finishes, executing agents.sh. From jdanigo/hydraia.
Hook
Runs when the agent finishes a response, executing summary.sh. From jdanigo/hydraia.