By trevoredris
Spec-driven development workflow: one plugin, seven commands, carrying work from discovery through TDD to review. PRD and roadmap; spec and architecture; slice-by-slice TDD; parallel code/security/a11y review; conventional commits; PR generation.
TDD execution. Slice-by-slice RED/GREEN/REFACTOR from a spec, with wave-parallel, debug, and refactor modes.
Interactive tour of fellowship. Detects project state and recommends a starting command. Accepts a task description to classify and recommend.
Bootstrap the .skilmarillion/projects/ scaffolding for a target project. Idempotent.
Spec-driven planning. Runs PRD to ROADMAP to specs pipeline, or individual stages via flags.
Parallel code-quality, security, and accessibility review. Deduplicated, prioritized findings. No code changes.
Accessibility specialist reviewer. WCAG 2.1/2.2 audit with severity ratings. Live browser verification when Playwright MCP is available; static analysis fallback. Findings only -- never proposes code changes.
Evaluate the spec against the codebase and recommend the appropriate architectural pattern.
Resolve a user's free-text / path / number input into a concrete artifact file (or ranked list of candidates) under `.skilmarillion/projects/*/`. Callers handle user confirmation via `AskUserQuestion`. This agent only discovers and ranks.
Code quality specialist reviewer. Evaluates architecture, design patterns, naming, complexity, duplication, and correctness. Findings only -- never proposes code changes.
Scan the codebase to identify relevant files, entry points, patterns, and conventions. Operates at two scopes:
Uses power tools
Uses Bash, Write, or Edit tools
Runs pre-commands
Contains inline bash commands via ! syntax
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.
Spec-driven development from discovery through TDD to review.
/plugin marketplace add TrevorEdris/skilmarillion
/plugin install fellowship@skilmarillion
| Command | Purpose | Model |
|---|---|---|
/fellowship:init | Bootstrap .skilmarillion/projects/{slug}/ scaffolding | haiku |
/fellowship:plan | PRD → roadmap → specs; --arch, --migrate, --validate | sonnet |
/fellowship:build | Slice-by-slice TDD; --debug and --refactor modes | sonnet |
/fellowship:review | Parallel code-quality / security / a11y; findings only | opus |
/fellowship:ship | Conventional commit; --pr opens a PR | haiku |
/fellowship:status | Read-only dashboard of active work | haiku |
/fellowship:help | Context-aware tour; recommends a command from a task description | haiku |
All artifacts land under the target project's git root at .skilmarillion/projects/{slug}/:
{target}/.skilmarillion/projects/{slug}/
PRD.md # /fellowship:plan --prd
ROADMAP.md # /fellowship:plan --roadmap (Phase → Wave → W{N}{letter})
DISCOVERY.md # /fellowship:plan --roadmap (inline context-gatherer, roadmap scope)
PROJECT-STATE.yaml # workflow state (current_wave, wave_agents_completed, …)
specs/SPEC-W{N}{letter}-{slug}.md # /fellowship:plan --specify (one per wave-agent)
adrs/NNN-{slug}.md # /fellowship:plan --arch adr
api/{name}-openapi.yaml # /fellowship:plan --arch api
schema/{name}-schema.sql # /fellowship:plan --arch schema
diagrams/{name}-{type}.md # /fellowship:plan --arch diagram
reviews/review-{target}.md # /fellowship:review
There is no plans/ directory — the SPEC absorbed the PLAN schema (ordered RED-GREEN-REFACTOR steps, files to touch, git strategy, traceability all live inside each SPEC).
.skilmarillion/ files are never auto-staged or auto-committed. The user decides whether to track them. /fellowship:ship ignores these paths unless explicitly asked.
Match model to task:
PRDs and SPECs are scored 0–100 by scripts/validate.py. PASS threshold: ≥85 for all document types. Draft threshold: 50.
.skilmarillion/
Commit the design artifacts the team agrees on. Keep per-engineer working state local.
.skilmarillion/projects/*/reviews/
.skilmarillion/projects/*/PROJECT-STATE.yaml
| Committed (shared) | Ignored (per-laptop) |
|---|---|
PRD.md, ROADMAP.md, DISCOVERY.md | PROJECT-STATE.yaml (resume state, current_wave, wave_agents_completed) |
specs/SPEC-W{id}-*.md | reviews/review-*.md (per-engineer findings) |
adrs/, api/, schema/, diagrams/ |
Rationale: SPECs, ADRs, and API/schema contracts are the team's durable agreement. SPECs may accumulate ACCEPT_WITH_DEBT annotations during /fellowship:build — that's part of the shared record, so they stay committed. Reviews are findings-only and feed the PR's review thread, not the repo.
Who runs which command depends on your role. Full playbook: HOW_TO_USE.md.
| Persona | Commands | Produces |
|---|---|---|
| Product Manager | /fellowship:plan --prd | PRD.md (validated ≥85) |
| Lead Engineer | /fellowship:plan --prd/--roadmap/--arch/--specify/--validate | DISCOVERY.md, ROADMAP.md (Phase → Wave → W{N}{letter}), specs/SPEC-W{id}-*.md (each ≥85), ADRs, API, schema, diagrams |
| Individual Engineer | /fellowship:build wave N or /fellowship:build spec W{id} → /fellowship:review → /fellowship:ship --pr | code + tests (one SPEC per wave-agent), PR with AC traceability |
Handoff: PM writes the PRD → Lead decomposes into a wave-based roadmap and PLAN-grade SPECs → engineers pick a wave (parallel) or a single SPEC and ship it.
MIT
npx claudepluginhub trevoredris/skilmarillionDeterministic Pokemon companion — roll a Pokemon from your user ID with ASCII sprites and shiny rendering
TrevorEdrisNASA Power of 10 rules for safety-critical code as agentic skills for Claude Code.
Silent session wiki: automatically captures and persists knowledge from every Claude Code session into a local markdown wiki — zero configuration required.
Harness-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
affaan-mComprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
JeffallanComprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
anthropicsTools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.
anthropicsComprehensive C4 architecture documentation workflow with bottom-up code analysis, component synthesis, container mapping, and context diagram generation
wshobsonAccess thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.
f