spec-driven-coding

Build software projects using structured spec-driven development with the Ralph Wiggum pattern. Use when asked to build, implement, or code a project, feature, or application. Covers discovery (requirements gathering), spec generation, AGENTS.md context generation, and autonomous implementation with fresh-context iterations. Supports any language/framework.

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

75%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-structured, mostly actionable skill body with clear phased workflow and good reference hygiene. The main gaps are a pseudocode orchestrator loop, unspecified error-recovery handling, and an unreferenced bundled script.

Suggestions

Replace the pseudocode orchestrator loop with executable guidance, or reference the bundled scripts/ralph-state.sh helper so the state-management commands are concrete and copy-pasteable.

Specify error-recovery behavior: what the orchestrator should do when .ralph-state.json has a non-empty last_error (retry, abort, or surface to the user) so the feedback loop is complete.

Trim motivational prose (e.g. "This is the key differentiator... Over time, they encode exactly how we build things.") to keep the body lean.

Dimension	Reasoning	Score
Conciseness	Mostly efficient with terse bullet lists and code blocks, but motivational lines like "This is the key differentiator: stacks improve with every project" and the "Why Ralph?" rationale are minor over-explanation that could be trimmed.	4 / 5
Actionability	Provides concrete, copy-pasteable artifacts (the .spec.json and .ralph-state.json schemas, the spawn prompt template, the stack file format), but the central orchestrator loop is explicitly labeled "# Pseudocode" rather than executable code.	4 / 5
Workflow Clarity	Four phases are clearly sequenced with checkpoints ("Confirm spec with user before proceeding", completion only when "everything builds and works", max-iteration/timeout limits), but error-recovery behavior for the state file's last_error field is left unspecified.	4 / 5
Progressive Disclosure	Clear overview with well-signaled one-level-deep links to three reference files (all verified present) plus a Quick Reference table, but the bundled scripts/ralph-state.sh helper is never surfaced or linked from the body.	4 / 5
	Total	16 / 20 Passed

Description

88%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description that clearly states both what the skill does and when to use it, with concrete named phases and natural trigger verbs. Minor room to broaden trigger synonyms and tighten distinctiveness against generic build/coding skills.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — "discovery (requirements gathering), spec generation, AGENTS.md context generation, and autonomous implementation with fresh-context iterations" — giving comprehensive coverage of the skill's capabilities.	5 / 5
Completeness	Explicitly answers both what ("Build software projects using structured spec-driven development...") and when ("Use when asked to build, implement, or code a project, feature, or application") with concrete trigger phrases.	5 / 5
Trigger Term Quality	Natural triggers like "build, implement, or code a project, feature, or application" are phrases users would actually say, but coverage stops there with no synonyms or variants beyond the core verbs/nouns.	4 / 5
Distinctiveness Conflict Risk	The "spec-driven development" + "Ralph Wiggum pattern" + "AGENTS.md" framing carves a clear niche, but the broad "build software projects" scope leaves minor overlap risk with general coding skills.	4 / 5
	Total	18 / 20 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: stukennedy/openclaw-starter-kit
Path: clawd/skills/spec-driven-coding/SKILL.md
Commit: 8d2a2a1

Reviewed: ~22 hours ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.