Tessl • The management layer for AI agent skills

SkillPublic

autoformat-input-rules

Check that every shorthand the www autoformat and basic marks kits declare applies whole when typed key by key, that no typed sequence of up to three keys changes a paragraph it was not typed in, that one undo takes back each keystroke that fires a rule, and that a key an input method commits gives what typing it gives. Use before and after changing apps/www/src/registry/components/editor/autoformat.tsx, basic-marks.tsx or another file in its coverage record, when the autoformat-rules step of pnpm check fails, or when fixing an autoformat or input-rule bug.

Quality

86%

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

77%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.

The body is a dense, fully actionable verification runbook: every command is copy-paste ready, the failure-triage workflow has explicit validation and feedback loops, and nothing wastes tokens on general knowledge. Its weaknesses are density and monolithic structure — bug-history narrative, a findings catalog, and a coverage table all sit inline in one long file with no progressive disclosure.

Suggestions

Move the Findings catalog, the Coverage record table, and the What a pass proves planted-bug details into a references/ file, keeping SKILL.md to the property, how to run, and how to read a failure, and link to them with clearly signaled one-level-deep references.

Tighten 'The harm' and 'What it checks' by splitting their long compound sentences; the prose often packs three or more facts per sentence, which costs the reader more parsing than the token savings are worth.

Add a brief run-order summary near the top (check → read failure → classify → fix → --lower → re-run) so the workflow sequence is visible before the dense detail sections.

Dimension	Reasoning	Score
Conciseness	Nothing explains concepts Claude already knows — every sentence carries project-specific detail — but sections like 'The harm' (bug-history narrative) and 'What a pass proves' (planted-bug count catalog) run long and could be tightened, matching the mostly-efficient-but-could-be-tightened anchor rather than the clearly verbose one.	3 / 5
Actionability	Fully executable guidance: copy-paste commands ('pnpm check:autoformat-rules', '--length=4', "--typed='<keys>'", '--jobs=<n>', '--lower', '--init'), concrete expected/got failure examples, and exact git commands for the coverage resync cover the common cases.	5 / 5
Workflow Clarity	The workflow has an explicit loop with validation checkpoints: run the gate, read a failure, classify real bug vs modeling error vs intended change before changing code, fix, then 'the gate fails until you run --lower'; the coverage resync names four confirmations and a gate run before moving the sync commit.	5 / 5
Progressive Disclosure	Headers are clear and sections well ordered, but the skill is a single 22 KB file with no bundle: the coverage-record table, the Findings catalog, and the What-a-pass-proves section are sizable reference material inlined in SKILL.md that could live in separate files, matching the some-structure-but-should-be-separate anchor.	3 / 5
	Total	16 / 20 Passed

Description

92%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 highly specific, well-triggered description: it states four concrete verification properties in third person and gives explicit use conditions tied to files, a failing check step, and bug fixes. Its only gap is missing a few natural synonyms (e.g. 'markdown shortcut', 'text substitution') that a user might say.

Dimension	Reasoning	Score
Specificity	The description lists four distinct concrete checks ('applies whole when typed key by key', 'no typed sequence of up to three keys changes a paragraph it was not typed in', 'one undo takes back each keystroke that fires a rule', 'a key an input method commits gives what typing it gives') in third person, matching the comprehensive-coverage anchor; nothing is generic enough to fall to 4.	5 / 5
Completeness	'Check that...' explicitly states the what, and 'Use before and after changing [files], when the autoformat-rules step of pnpm check fails, or when fixing an autoformat or input-rule bug' explicitly states the when with concrete triggers, matching the both-what-and-when anchor exactly.	5 / 5
Trigger Term Quality	Natural terms like 'autoformat', 'input rule'/'input-rule bug', 'input method', 'pnpm check fails' plus exact file paths give good keyword coverage, but common synonyms a user might say ('markdown shortcut', 'text substitution') are missing, so it sits at the good-coverage anchor rather than comprehensive.	4 / 5
Distinctiveness Conflict Risk	Repo-specific kit names, exact file paths, and a named check step ('autoformat-rules step of pnpm check') carve a clear niche with minimal risk of triggering for a wrong skill.	5 / 5
	Total	19 / 20 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	13 / 16 Passed

Repository: udecode/plate
Path: .agents/skills/autoformat-input-rules/SKILL.md
Revision: f0c616a

Reviewed: 3 days ago

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.

autoformat-input-rules

Table of Contents

Is this your skill?