writing-skills

Use when creating new skills, editing existing skills, or verifying skills work before deployment

Quality

77%

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

Fix and improve this skill with Tessl

tessl review fix ./plugins/superpowers/skills/writing-skills/SKILL.md

The canonical home for this skill is writing-skills in obra/superpowers

Quality

Content

55%

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

The body is highly actionable with clear, validated TDD workflows, but it is severely over-long and self-contradictory: it violates its own token-efficiency budget through heavy repetition and uses `@` references to non-existent bundle files despite explicitly advising against that syntax.

Suggestions

Trim the body toward the skill's own <500-word target by collapsing the repeated RED-GREEN-REFACTOR, Iron Law, and rationalization-table content into single canonical sections and cross-referencing them.

Remove or replace `@`-prefixed references (`@testing-skills-with-subagents.md`, `@graphviz-conventions.dot`) with the skill-name-only style it recommends, and either add the referenced bundle files or drop the references.

Split the heavy testing-methodology and CSO material into separate reference files so SKILL.md acts as a lean overview, demonstrating the progressive-disclosure structure the skill teaches.

Dimension	Reasoning	Score
Conciseness	The ~650-line body vastly exceeds the skill's own "Other skills: <500 words" target and repeats material across sections (TDD mapping, rationalization tables, the Iron Law no-exceptions list, and RED-GREEN-REFACTOR each appear multiple times), padding rather than trusting Claude's intelligence.	1 / 3
Actionability	It provides copy-paste-ready templates (rationalization table, red-flags list), concrete commands (`wc -w`, `render-graphs.js`), explicit before/after examples, and a per-item checklist — fully executable guidance.	3 / 3
Workflow Clarity	The RED-GREEN-REFACTOR cycle is explicitly sequenced with a TDD mapping table and verification checkpoints ("Run scenarios WITH skill - verify agents now comply", "Re-test until bulletproof"), plus a mandatory deployment checklist with feedback loops.	3 / 3
Progressive Disclosure	No bundle files exist, yet the body references files that are missing (`@graphviz-conventions.dot`, `@testing-skills-with-subagents.md`, `anthropic-best-practices.md`, `persuasion-principles.md`) and even uses the `@` force-load syntax the skill itself warns against; all content is crammed into one monolithic file.	1 / 3
	Total	8 / 12 Passed

Description

100%

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

The description is concise, third-person, and leads with an explicit "Use when..." trigger naming three concrete skill-authoring actions. It cleanly answers both what and when without summarizing workflow, consistent with the skill's own CSO guidance.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — "creating new skills, editing existing skills, or verifying skills work before deployment" — matching the anchor for specific concrete actions rather than a single vague verb.	3 / 3
Completeness	Opens with an explicit "Use when..." trigger and the named actions convey what the skill covers, satisfying both the what and when with explicit triggers.	3 / 3
Trigger Term Quality	Phrases like "creating new skills", "editing existing skills", and "verifying skills" are exactly the natural terms a user would say when needing this skill, with good synonym coverage.	3 / 3
Distinctiveness Conflict Risk	The skill-authoring niche is clearly bounded with distinct triggers, making overlap with unrelated skills unlikely.	3 / 3
	Total	12 / 12 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (656 lines); consider splitting into references/ and linking	Warning

	Total	15 / 16 Passed

Repository: openai/plugins
Path: plugins/superpowers/skills/writing-skills/SKILL.md
Commit: 11c74d6

Reviewed: about 10 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.