Assess Spec — Testing skill for Claude Code
Analyze an unbuilt spec for implementability risks - oversized stories, deep dependency chains, file-overlap conflicts - and recommend decomposition.
How to install Assess Spec
Installs to ~/.claude/skills/sellke-writ-assess-spec/SKILL.md
mkdir -p ~/.claude/skills/sellke-writ-assess-spec && curl -fsSL https://raw.githubusercontent.com/sellke/writ/HEAD/commands/assess-spec.md -o ~/.claude/skills/sellke-writ-assess-spec/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Assess Spec does
name: assess-spec description: "Analyze an unbuilt spec for implementability risks - oversized stories, deep dependency chains, file-overlap conflicts - and recommend decomposition. Run when a spec feels too heavy to build." problem: "An oversized or deeply chained spec reads fine on paper and only fails once /implement-spec is halfway through it, with orchestrator context already spent." outcome: "The unbuilt spec has a stated implementability rating backed by all six checks, with every fla
Alternatives in Testing
- Test Scenarios — Create comprehensive test scenarios from user stories 7.8k ★
- Spec Driven Develop — Spec-driven development workflow for AI coding agents: architecture-first planning, task decomposition, GitHub 975 ★
- Mda — MDA Open Spec — a Markdown superset for agent-facing documents 615 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Flawz
Hostile-but-honest flaw hunt on a plan, spec, design, or reasoning (not code diffs — that's /flow:deep-review)
Spec Validate
Analyze a specification document to determine if it has enough detail for autonomous implementation. Use to va
Amby.Analyze
Cross-check spec, plan, and tasks for coverage, consistency, and dependency-graph health.
Glm Delegate
Delegate token-heavy execution to swarms of GLM agents in disposable git worktrees — an Anthropic model writes
Spectastic.Design
Produce the design doc for an existing spec — stack, architecture, decisions, risks, ADRs. Use when deciding h
Unity Refine
Refines a raw GitHub issue into an implementation-ready spec — problem, acceptance criteria, scope, affected s
Related Agents
Fixme Review Spec
Reviews specifications for deterministic implementability. Read-only - produces evidence-backed findings about
Land Analyst
Analyzes the 8 land factors: contiguous developable area, slope, price, zoning, protected overlap, brownfield
Ticket Scoper
Given a request description, produces a complete correctly-formatted ticket and flags any conflicts (duplicate