writing-guide

Use when writing or editing any Spacebot copy — README sections, docs, release notes, marketing text, design doc summaries. Covers voice, tone, patterns to avoid, and what good Spacebot writing sounds like.

Quality

88%

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

83%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.

An unusually crisp style guide: dense bad/good example pairs make the rules directly executable, and there is no filler. The weaknesses are structural rather than substantive — it reads as a rule reference with no application workflow or final-pass checklist, and all content sits inline in one file.

Suggestions

Add a short 'How to apply' workflow at the top: draft/edit the copy → re-read the draft scanning for each pattern in 'Patterns to Avoid' → check 'Words to Avoid' → confirm claims match the differentiator list. This gives an explicit sequence with a verification checkpoint.

Include a final-pass checklist (e.g., 'Before finishing: no em dashes in sentences, no "Not X. Not Y." openers, no banned words') so Claude can validate its own output against the guide.

Provide good rewrites for the two patterns that currently only say "Pick one" and "Just cut it" ('Not only X, but also Y' and 'No X. No Y.' closers), matching the Bad→Good format used for every other pattern.

Dimension	Reasoning	Score
Conciseness	Lean throughout: every pattern is a one-line rule plus a tight bad/good pair, with no explanation of concepts Claude already knows — e.g., "Semicolons in prose. Use a period." followed directly by the example. Every token earns its place.	5 / 5
Actionability	Highly actionable for an instruction-only skill: each anti-pattern has a concrete Bad→Good rewrite pair (e.g., "Bad: 'The cortex sees across all channels — the only process with full system scope.' / Good: 'The cortex is the only process that sees across all channels.'), plus an explicit word-avoid list. Only two patterns ('Not only X, but also Y', 'No X. No Y.' closers) give an instruction ("Pick one", "Just cut it") without a rewrite — a minor gap.	5 / 5
Workflow Clarity	The guide is a reference of rules, not a sequenced procedure: there is no explicit order for applying the rules when writing or editing (e.g., draft → scan against pattern list → check word list), and no verification/checklist step to confirm a draft passes. Sections flow logically, but the anchors at 4–5 require a clear sequence with checkpoints, which is absent.	3 / 5
Progressive Disclosure	Well-organized single-file skill with clear section headers and no bundle files; navigation within the file is easy. However, at ~105 lines everything is inline in SKILL.md with no overview-then-reference split (e.g., the pattern list or word-avoid list could live in a reference file), so it falls just short of the anchor-5 structure.	4 / 5
	Total	17 / 20 Passed

Description

87%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: explicit "Use when..." trigger clause with a concrete list of artifact types, a clear statement of what the skill covers, third-person voice, and a distinct niche. The only gaps are a thin action vocabulary (write/edit) and a few missing natural synonyms.

Dimension	Reasoning	Score
Specificity	Names concrete targets ("README sections, docs, release notes, marketing text, design doc summaries") and concrete coverage areas ("voice, tone, patterns to avoid"), but the actions themselves are limited to "writing or editing" — minor gap versus the anchor-5 example's multiple distinct actions.	4 / 5
Completeness	Explicitly answers both: "Use when writing or editing any Spacebot copy — README sections, docs, release notes, marketing text, design doc summaries" (when, with concrete triggers) and "Covers voice, tone, patterns to avoid, and what good Spacebot writing sounds like" (what).	5 / 5
Trigger Term Quality	Good natural keywords users would actually say ("README", "docs", "release notes", "marketing text", "design doc"), but common variations like "documentation", "blog post", or "copywriting" are missing.	4 / 5
Distinctiveness Conflict Risk	Scoped to a clear niche ("any Spacebot copy") with distinct triggers tied to specific artifact types; minimal overlap risk with generic writing or documentation skills.	5 / 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: spacedriveapp/spacebot
Path: .agents/skills/writing-guide/SKILL.md
Commit: 5c2ee47

Reviewed: 9 days 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.