format-review-comments
Skill4 starsChanged 6 months agoMITView on GitHub
Formats code review and PR feedback using the Conventional Comments standard (label [decorations]: subject). Apply proactively whenever posting review comments or any critique of code or pull requests. Invoke directly when the user asks to format feedback, use conventional comments style, or label a review comment as blocking, suggestion, or nitpick.
In this repository13 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- audit-skillsSkill
- create-commitSkill
- implement-planSkill
- refine-proseSkill
- review-agents-mdSkill
- review-planSkill
- review-prSkill
- review-skillSkill
- upsert-agents-mdSkill
- upsert-planSkill
- upsert-skillSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 2 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 26
Use this format whenever you are reviewing anything — no explicit request needed.
- Distinctive: line 30
When invoked directly to format existing comments:
- Very specific: line 33
Read references/conventional-comments-spec.md to confirm label and decoration meanings.
- Very specific: line 56
Read references/conventional-comments-spec.md before selecting labels for any comment that is not obviously a nitpick or praise.
- Very specific: line 70
When selecting a label for an existing comment, match the comment's intent to the label definitions in …
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Conventional Comments
Formats review comments using the Conventional Comments standard, making intent unambiguous — readers know immediately whether a comment is blocking, optional, or informational.
When to apply
Apply this format proactively whenever you:
- Post a PR review comment (inline or summary)
- Provide feedback on code, architecture, or a document
- Suggest a change or improvement in a review context
Use this format whenever you are reviewing anything — no explicit request needed.
Steps
When invoked directly to format existing comments:
- Identify the comments to format — from context, the user's message, or by asking if unclear.
- Read
references/conventional-comments-spec.mdto confirm label and decoration meanings. - For each comment, choose the label that best matches the comment's intent (see Core labels and label-selection guidance below).
- Apply the format:
label [decorations]: subjectfollowed by an optional body on a new line. - Output all reformatted comments in the same order as the originals.
Format
label [decorations]: subject
[body]
[discussion reference]- label — the comment's intent (see reference for full list)
- decorations — optional modifiers in brackets, comma-separated (e.g.
[blocking, if-minor]) - subject — a short summary; the main message
- body — optional elaboration, reasoning, or examples
- discussion reference — optional link to a related issue, PR, or discussion
Core labels
Read references/conventional-comments-spec.md before selecting labels for any comment that is not obviously a nitpick or praise.
| Label | Intent |
praise | Highlights something positive; never blocking |
nitpick | Minor style/preference issue; non-blocking by default |
suggestion | Proposes an improvement; non-blocking by default |
issue | Points out a problem that must be addressed; blocking by default |
todo | Small, unambiguous task the author should complete |
question | Asks for clarification; non-blocking |
thought | Shares an idea that doesn't warrant action |
chore | Simple housekeeping (rename, move, reformat); non-blocking |
note | Informational; no action required |
When selecting a label for an existing comment, match the comment's intent to the label definitions in references/conventional-comments-spec.md. When intent is ambiguous, prefer suggestion over issue and note over thought.
Quick examples
praise: clean abstraction here — easy to follow
suggestion: consider extracting this into a helper so it can be reused
issue [blocking]: this will panic on nil input; add a nil check before dereferencing
question: why is this sorted descending? just want to make sure it's intentional
nitpick: s/recieve/receive/Example with body:
suggestion: extract the retry logic into a helper
This block appears in three places. A shared helper would make it easier to adjust the backoff strategy in one place.Direct invocation — before and after:
Input:
This will crash on empty lists. Also the variable name is confusing.
Output:
issue [blocking]: this will crash on empty lists — add a guard before iterating
nitpick: variable name is unclear — consider renaming to reflect its purposeDoes it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

