By volodchenkov
SDLC workflow agents and skills for Plane Conductor — BABOK, C4, DDD, ISTQB, OWASP, WCAG-grounded.
API Tester agent. Use when backend code (CHANGES from a backend developer) is ready and the REST API needs system-level testing — endpoint behaviour, status codes, idempotency, multitenancy, performance smoke. Designs test cases per ISTQB Foundation framework.
Architect agent. Use when SPEC is ready by the system-analyst and needs an ARCH_REVIEW — service boundaries, multitenancy, performance, transactions, integration security, migrations, ADR governance, traceability validation. Produces verdict APPROVED / CHANGES_REQUIRED.
Business Analyst agent. Use when a new product task arrives in Plane and requirements need to be elicited from the initiator and structured into the root issue description per BABOK v3 framework.
UX/UI Designer agent. Two modes — Mode A (Design brief): produce UX flow + Figma frame inventory before frontend coding. Mode B (UX review): verify frontend implementation matches design intent. Follows Nielsen heuristics + WCAG 2.1 + platform conventions.
Django/DRF backend developer. Use when SPEC is approved by the architect and Django code needs to be written — models, views, serializers, Celery tasks, migrations.
Use this skill when working as a Business Analyst (the business-analyst role) — eliciting requirements from stakeholders, structuring requirements per BABOK v3 framework, classifying needs by type (Business / Stakeholder / Solution / Transition), and applying analysis techniques (5 Whys, MoSCoW, INVEST, Stakeholder Analysis). Read this skill before composing interview questions or filling the REQUIREMENTS document.
Use this skill when working as the final Reviewer (the reviewer role) — comprehensive cross-cutting review before closing a Plane issue. Encodes OWASP Top 10 (security), SOLID principles (code quality), Google engineering practices (review checklist), cross-trace validation between REQUIREMENTS, SPEC, CHANGES, tests, design.
Use this skill when working as a coder (the django-developer / the vue-developer / the react-developer) and your implementation is complete — what documentation to update, where, and how. Covers docstrings, README updates, API endpoint docs, ADR status updates, migration notes, inline comments. The author of the change is responsible for the documentation.
Use this skill when working as a tester (the api-tester for API testing, the ui-tester for UX/E2E testing) — designing test cases per ISTQB Foundation framework, choosing the right test design technique (Equivalence Partitioning, Boundary Value Analysis, Decision Table, State Transition, Use Case Testing), classifying test levels and types, structuring bug reports. Read before composing a test plan or executing test cases.
Use this skill whenever an agent interacts with Plane (https://plane.so) — picking up issues, creating sub-issues for artifacts, posting comments, finding artifacts by label, mentioning users, handling re-entry (continuation vs rework), or completing work. Triggers on any usage of mcp__plane__* tools. Encodes the Plane Conductor pipeline protocol.
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
SDLC agent pack for Plane Conductor — 10 pipeline agents + 2 meta-agents, grounded in industry methodologies (BABOK, C4, ISTQB, OWASP, WCAG).
A starter pack of specialised Claude Code agents wired to follow standard software-engineering methodologies, designed to drop into Plane Conductor as a full SDLC pipeline triggered by Plane mentions.
Project-specific details (stack, conventions, multi-tenancy rules,
verification commands, frontends inventory) are kept outside the
prompts — in your own project's KB, populated from the template/
directory in this repo.
agents/ — generic agent definitionsFile (also the name: for claude --agent <name>) | Role | Methodology |
|---|---|---|
business-analyst.md | Business Analyst | BABOK v3 (4 requirement types, MoSCoW, INVEST, 5 Whys) |
system-analyst.md | System Analyst | C4 + DDD bounded contexts + REST + IEEE 29148 + ADR |
architect.md | Architect | SOLID + DDD + ADR governance + 6-area review checklist |
designer.md | UX/UI Designer | WCAG 2.1 AA + Nielsen heuristics + Material/HIG |
django-developer.md | Django backend developer | project-conventions-driven |
vue-developer.md | Vue developer (Vue 2 / Vue 3) | project-conventions-driven |
react-developer.md | React / Next.js developer | project-conventions-driven |
api-tester.md | API Tester | ISTQB Foundation (EP, BVA, decision tables, state transition, use-case) |
ui-tester.md | UX/E2E Tester | ISTQB + WCAG 2.1 AA |
reviewer.md | Final Reviewer | OWASP Top 10 + SOLID + Google review checklist |
| File | name: (alias) | What it does |
|---|---|---|
prompt-architect.md | zuse | Designs and audits agent prompts, skills, and KB reference docs. Plan-then-edit; runs a 9-block audit on every prompt before changing it. |
project-manager.md | tron | Personal PM. Triages incoming work into "fix it myself" / "file in Plane and run the pipeline" / "clarify first". Reads GitHub / GitLab / kubectl / helm freely; every state change requires explicit approval. |
Each file is a Claude Code agent definition
with YAML frontmatter (name, description, model, tools) and a
long-form prompt body. SDLC prompts read project-specific KB files (see
below) and the matching skill before composing artifacts. Meta-agent
prompts have no project-specific reads — they work in any repo.
skills/ — 9 reusable Claude Code skills| Skill | Used by | Encodes |
|---|---|---|
babok-elicitation | business-analyst | BABOK v3 elicitation + analysis knowledge areas |
system-design-techniques | system-analyst | C4, DDD bounded contexts, REST conventions, ADR pattern |
architecture-review-framework | architect | 6-area review checklist, SOLID + DDD lenses, ADR governance |
ux-design-discipline | designer | WCAG 2.1 AA, Nielsen heuristics, platform conventions |
documentation-discipline | coders | docstrings, README updates, ADR status, migration notes |
istqb-test-design | api-tester, ui-tester | EP / BVA / decision tables / state transition / use case |
code-review-discipline | reviewer | OWASP Top 10, SOLID, Google review checklist, cross-trace |
artifact-templates | every agent | canonical Markdown templates for every artifact type |
plane-operations | every agent | the Plane read/write protocol (pickup, sub-issue, comments, mentions, re-entry) |
template/ — KB template you copy into your projectProject-specific facts go in your own repo — versioned alongside your code. The pack ships an empty template:
template/
├── AGENTS.md # entry point: index, routing table, project rules at a glance
├── PROJECT_DESCRIPTION.md # draft of the Plane project description (operational layer)
└── kb/
├── stack.md # languages, frameworks, libs, versions
├── conventions.md # lint, types, naming, TDD policy
├── architecture.md # services / bounded contexts / import contracts
├── verify.md # slash-commands / make / npm scripts for lint, tests, builds
├── multitenancy.md # tenant isolation rules (or "N/A")
├── migrate.md # DB migration discipline
├── frontends.md # per-frontend stacks, build commands
├── document.md # docstring style, doc-generation tool
└── domain/ # ad-hoc domain knowledge files
You copy template/ into your repo, fill it in, commit. The agents
read it via $KB_DIR env var (set by Plane Conductor) or fall back to
<cwd>/AGENTS.md.
npx claudepluginhub volodchenkov/claude-sdlc-agents --plugin sdlc-agentsHarness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
v9.44.1 — Patch release for Gemini environment/version detection and qwen auth gating. Run /octo:setup.
AI-powered development tools for code review, research, design, and workflow automation.
Superpowers Plus core skills library for Claude Code: planning, execution routing, TDD, debugging, and collaboration workflows
Unity Development Toolkit - Expert agents for scripting/refactoring/optimization, script templates, and Agent Skills for Unity C# development
Claude harness - A harness for solo developers (Vibecoders) to handle full-cycle contract development.