Skip to content
SkillModel-led2,104 tokens in use

Grammar Check

Finds grammar, logic and flow errors in a text, judged against its stated objective, and lists targeted fixes with reasons without rewriting the whole piece.

28,201 stars on the repository MIT Licensephuryn/pm-skills · pm-toolkit/skills/grammar-check
claude plugin marketplace add phuryn/pm-skills claude plugin install pm-toolkit@pm-skills
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing Grammar Check costs

39tokens idle
2,104tokens in use

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Counts are approximate, made offline with cl100k_base.

Model dependence Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

Nothing detected: no shell, network, file writes, git or credentials in its files.

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Excellent
  • Frontmatter12/15
  • Structure25/25
  • Examples3/15
  • Output15/15
  • Size15/15
  • Resources10/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent

No findings.

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Good
  • Development activity80
  • Maintenance76
  • Community60
  • Maturity32
  • Momentum40

Grammar Check reviews a text you provide together with its objective, such as persuading investors or explaining features to new users. The agent notes the likely audience and tone, then scans for grammar errors, logical errors and flow errors.

Each finding comes with its location, the error, a suggested fix and the reason it matters. The agent does not return a corrected full text, so the author keeps their own voice, and it leaves alone intentional style choices and correct informal language.

What You Get

  • An error summary with counts of grammar, logic and flow errors
  • A list of fixes by category, each with location, error, fix and reason
  • A shortlist of the three to five changes with the largest effect
  • A short assessment of how well the text meets its objective and whether the tone fits

Common Use Cases

  • Proofreading a product document, email or presentation
  • Checking that marketing copy supports its goal
  • Finding unsupported claims and vague statements in a draft
  • Reviewing a teammate’s writing without changing their voice

What Makes It Unique

The skill returns specific fix suggestions and does not rewrite the document, and it judges the text against the objective you state, which adds a logic check for unsupported claims, contradictions and vague promises beyond spelling and punctuation. It includes examples of weak and strong feedback and a list of cases where no change is the right answer, such as deliberate short sentences.

Under The Hood

The skill is a single prompt file with no scripts. The agent loads the instructions, the error categories with examples and the review checklist when the skill is used, then reports in the conversation on the text you paste. It does not call outside services, run commands or write files by itself.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers