Engineering plugins

Closed-loop scaffolder for unattended agent operations: oracle scoring, headless agent loops, hooks, and worktree races — tailored to each project's signals.

12 skills · 2 agents

Workflow routing, review choices, interactive grilling, and Thai commit messages.

14 skills

Project knowledge for the noyalib YAML library: its architecture contract, build, CI and release, debugging and validation. Install it only if you work on noyalib.

14 skills

Development pipeline skills (explore, plan, implement, test, cleanup, review, resolve) and focused skills such as anti-slop-code.

11 skills · 2 agents

Orchestrate plans, workflows, implementation tracking, and reviews — import plans from Claude sessions, create XML-enhanced workflows, review implementations, and track progress

11 skills · hooks

KiCad EDA automation for AI agents through kicad-mcp: schematic, PCB, DRC, validation, and fabrication workflows.

11 skills · 1 MCP server

Skills for disciplined day-to-day engineering: phased planning, vertical-slice builds with review checkpoints, TDD, debugging, and code review.

10 skills

See where a project stands, set it up by interviewing the engineer, read and write requirements, turn specs and designs into requirements, prove them with tests, review diffs against them, and run design review gates.

9 skills

The engine: the always-loaded kernel, the operating doctrine, the universal skills, and the personas as agents — shipped with the Workspace Definition and a complete demo workspace to read.

4 skills · 5 agents

Apply expert thinking profiles to code reviews, architecture evaluation, and security audits - think like DHH, Sandi Metz, or build your own expert profiles

5 skills · 2 agents · hooks

Operate the self-hosted Prefect server: access, HeavyPad workers, deployments, ad-hoc runs, and infrastructure costs.

5 skills · 3 MCP servers

OOAD + Clean 4 Layers + TDD workflow: setup, discover, requirements, architect, build, verify, ship, plus ask-ooad router.

8 skills

Ownership-preserving feature workflow, complexity budgets, Out-of-the-Tar-Pit state rules, evidence-based code review and codebase knowledge maps.

8 skills

Vic Sage investigation discipline: interrogate specs, investigate root causes with a provenance-backed case board, write sourced reports, faceless voice.

6 skills

Develop comprehensive, ISO 55001:2024-aligned Asset Management Plans through adaptive interviews, data analysis, and autonomous research. Produces both interactive HTML presentations and branded DOCX documents. Sector-agnostic with LCC analysis, D3.js charts, and per-section quality review.

1 skill · 1 command · 4 agents

Language-agnostic engineering standards as triggered skills: frontend, backend, database, security and git.

6 skills

Conventions that fix themselves — evidence-gated convention repair loop plus a budget/archive system for bloated rule docs. Stack-agnostic (FE/BE/scripts/docs).

3 skills · 1 agent · hooks

EasyEDA automation for AI agents through easyeda-mcp-pro: project workflows, component search, design checks, and EDA assistance.

4 skills · 1 MCP server

Structured implementation planning, interactive annotation review via Plannotator (with a local fallback), and plan execution with an adaptive review cascade sized to the plan.

2 skills · 1 command · 1 agent · hooks

Accessibility + AEO inspection for Claude Code. Lighthouse-style scoring, interactive reports, jurisdiction-aware WCAG context, and auto-advisory hook.

3 commands · hooks

DSI RSE agent skills: structured PR review, supervised model search, clinic project review, and more.

4 skills

Adyan's agent skills for everyday engineering (before/after demo videos, UI/UX revamps and more).

4 skills

go-deeper（把 prompt 改寫得更深）、overnight（睡覺時讓 agent 顧 PR）、hotspot-review（找最近常被修的地方並重構）、decision-form（把待決問題做成手機問卷，答完複製貼回）。

4 skills

Agent workflow, context engineering, and primary-source research skills.

3 skills

Carter's agent skills — fix-don't-freeze CI discipline, worker reasoning checks, and the j-space reasoning suite.

3 skills

Human-verification gate for an agent loop: builds a single-file review page (the page itself makes no network request; a reviewed HTML artifact's own https: assets still load), collects batched feedback as structured batch.v1 data instead of chat prose, and refuses to close while a BLOCKER is open, the reviewer is unnamed, or nobody has reviewed. Non-blocking, headless-guarded, round-capped. Stdlib-only.

1 skill · 1 command · 1 agent

glendonize, grilling and whiteboard: make a repo agent-ready, stress-test a plan, draw and be read.

3 skills

Grade a change the way an external reviewer will, before pushing it.

1 skill · 1 agent

Practical Python project setup, modern uv/ruff/ty/pytest workflows, PEP 723 scripts, and careful migration from legacy Python tooling.

1 skill · hooks

Observability for your coding agent sessions. Session and commit telemetry, synced to gjalla cloud automatically.

1 MCP server · hooks