Production-grade engineering skills for AI coding agents.
Built by Systango
AI coding agents default to the shortest path — which often means skipping specs, tests, security reviews, and the practices that make software reliable. At Systango, we built Agent Skills to give agents structured workflows that enforce the same discipline our engineers bring to production code.
Each skill encodes hard-won engineering judgment: when to write a spec, what to test, how to review, and when to ship. These aren't generic prompts — they're opinionated, process-driven workflows that separate production-quality work from prototype-quality work.
The skills draw on well-established principles: Hyrum's Law in API design, change sizing in code review, Chesterton's Fence in simplification, trunk-based development, Shift Left and feature flags in shipping, and deprecation patterns that treat code as a liability — embedded directly into step-by-step workflows agents follow.
| Tool | Get started |
|---|---|
| Claude Code | /plugin install sys@systango-agent-skills |
| Cursor | curl -fsSL https://raw.githubusercontent.com/SystangoTechnologies/agent-skill-library/main/scripts/install-cursor.sh | bash → commands use /sys-* |
| Gemini | gemini skills install https://github.com/SystangoTechnologies/agent-skill-library.git --path skills |
Step-by-step install, SSH workarounds, uninstall, and per-tool setup: INSTALLATION.md · Getting started · Cursor · Gemini
- Why Agent Skills?
- Quick start
- Lifecycle flows
- Commands
- Skills
- Agents & references
- How skills work
- Project structure
- License
Modular agent skills across the SDLC. Cursor installs use the /sys-* prefix shown on the diagrams below; Claude Code uses the same names without the prefix (for example /sys-spec → /spec).
Core engineering flow from codebase discovery through ship. Mermaid diagrams →
PM-first path: PRD → epic → breakdown → Jira sync, then hand off to engineering. Mermaid diagrams →
Jira-tracked engineering (story-grounded spec, sync.md preview before board updates) is documented in docs/lifecycles.md.
17 slash commands map to the development lifecycle. Each activates the right skills automatically.
| What you're doing | Command | Key principle |
|---|---|---|
| Discover a codebase | /discover |
Understand before you build |
| Define what to build | /spec |
Spec before code |
| Plan how to build it | /plan |
Small, atomic tasks |
| Sync Jira after plan | /jira-sync |
sync.md diff preview; push to Jira only after approval |
| Build incrementally | /build |
One slice at a time |
| Prove it works | /test |
Tests are proof |
| Review before merge | /review |
Improve code health |
| Simplify the code | /code-simplify |
Clarity over cleverness |
| Ship to production | /ship |
Faster is safer |
| Small, ad-hoc change | /quick |
Tracked, atomic, fast |
| Define from Jira | /jira-spec |
Jira-grounded spec |
| Plan from Jira | /jira-plan |
Story-scoped task plans |
| Ship with Jira sync | /jira-ship |
Launch + ticket updates |
| PM PRD creation | /pm-spec |
SysPM PRD-first planning |
| PM PRD to epic | /pm-epic |
Convert PRD into epic artifact |
| PM epic breakdown | /pm-breakdown |
Decompose epic into T### work items |
| PM Jira sync | /pm-sync |
Sync SysPM epic and work items to Jira |
Skills also activate automatically based on what you're doing — designing an API triggers api-and-interface-design, building UI triggers frontend-ui-engineering, and so on.
31 skills across the SDLC — core lifecycle, Jira-aware variants, SysPM PM workflows, and supporting skills. Each is a structured process with steps, verification gates, and anti-rationalization tables.
| Phase | Count | Examples |
|---|---|---|
| SysPM | 5 | pm-spec-and-prd, syspm, pm-jira-sync |
| Discover | 1 | brownfield-discovery |
| Define | 3 | spec-driven-development, jira-spec-driven-development |
| Plan | 2 | planning-and-task-breakdown, jira-planning-and-task-breakdown |
| Sync Jira | 1 | jira-board-sync |
| Build | 6 | incremental-implementation, frontend-ui-engineering |
| Verify | 2 | debugging-and-error-recovery, browser-testing-with-devtools |
| Review | 4 | code-review-and-quality, security-and-hardening |
| Ship | 5 | shipping-and-launch, git-workflow-and-versioning |
| Ad-hoc | 1 | quick-task-execution |
| Meta | 1 | using-agent-skills |
Full catalog with descriptions →
| Resource | Description |
|---|---|
| Agent personas | Code reviewer, test engineer, security auditor |
| Reference checklists | Testing, security, performance, accessibility patterns |
Every skill follows a consistent anatomy:
┌─────────────────────────────────────────────┐
│ SKILL.md │
│ │
│ ┌─ Frontmatter ─────────────────────────┐ │
│ │ name: lowercase-hyphen-name │ │
│ │ description: Use when [trigger] │ │
│ └───────────────────────────────────────┘ │
│ │
│ Overview → What this skill does │
│ When to Use → Triggering conditions │
│ Process → Step-by-step workflow │
│ Rationalizations → Excuses + rebuttals │
│ Red Flags → Signs something's wrong │
│ Verification → Evidence requirements │
└─────────────────────────────────────────────┘
Key design choices:
- Process, not prose. Skills are workflows agents follow, not reference docs they read. Each has steps, checkpoints, and exit criteria.
- Anti-rationalization. Every skill includes a table of common excuses agents use to skip steps (e.g., "I'll add tests later") with documented counter-arguments.
- Verification is non-negotiable. Every skill ends with evidence requirements — tests passing, build output, runtime data. "Seems right" is never sufficient.
- Progressive disclosure. The
SKILL.mdis the entry point. Supporting references load only when needed, keeping token usage minimal.
agent-skill-library/
├── assets/ # Banner and lifecycle diagrams
├── skills/ # 31 skills (SKILL.md per directory)
│ ├── brownfield-discovery/ # Discover
│ ├── idea-refine/ # Define
│ ├── spec-driven-development/ # Define
│ ├── jira-spec-driven-development/ # Define (Jira)
│ ├── planning-and-task-breakdown/ # Plan
│ ├── jira-planning-and-task-breakdown/ # Plan (Jira)
│ ├── jira-board-sync/ # Sync Jira (after plan)
│ ├── incremental-implementation/ # Build
│ ├── generate-unit-tests/ # Verify (test generation)
│ ├── context-engineering/ # Build
│ ├── source-driven-development/ # Build
│ ├── frontend-ui-engineering/ # Build
│ ├── api-and-interface-design/ # Build
│ ├── browser-testing-with-devtools/ # Verify
│ ├── debugging-and-error-recovery/ # Verify
│ ├── code-review-and-quality/ # Review
│ ├── code-simplification/ # Review
│ ├── security-and-hardening/ # Review
│ ├── performance-optimization/ # Review
│ ├── git-workflow-and-versioning/ # Ship
│ ├── deprecation-and-migration/ # Ship
│ ├── documentation-and-adrs/ # Ship
│ ├── shipping-and-launch/ # Ship
│ ├── jira-shipping-and-launch/ # Ship (Jira)
│ ├── pm-spec-and-prd/ # SysPM
│ ├── pm-epic-decomposition/ # SysPM
│ ├── pm-jira-sync/ # SysPM
│ ├── pm-status-and-reporting/ # SysPM
│ ├── syspm/ # SysPM mirrored PM workflow
│ ├── quick-task-execution/ # Ad-hoc
│ └── using-agent-skills/ # Meta
├── agents/ # 3 specialist personas
├── references/ # Supplementary checklists
├── hooks/ # Session lifecycle hooks
├── .claude/commands/ # 17 slash commands
├── INSTALLATION.md # Install guide (all tools)
├── SKILLS.md # Full skill catalog
└── docs/ # Lifecycles, setup guides, agents, references
MIT — use these skills in your projects, teams, and tools. See LICENSE.
Systango · Install · Skills · Lifecycles · Getting started · Agents · References