agentleFS
imMamdouhaboammarimMamdouhaboammar/anti-slop-engineering/skills/comment-budget-guard/SKILL.md

comment-budget-guard

Skill4 starsChanged 31 days agoMITView on GitHub

Enforce Comment-to-Code Ratios (CCR) and eliminate syntax play-by-play narration comments while strictly preserving formal safety contracts, JSDoc/TSDoc docstrings, and license headers.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Silence over noise: If a line of code is obvious to a competent engineer, narrating it with comments is banned.

    line 26
  2. Very specific:

    Contract preservation: Never strip or discourage // SAFETY: proof annotations, formal docstrings, or license headers.

    line 27

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Comment Budget & Obviousness Guard

Enforce the Zero-Obviousness & Anti-Truism Doctrine by bounding Comment-to-Code Ratios (CCR) and eliminating syntax play-by-play comments without penalizing docstrings or safety invariants.

Why This Skill Is Needed & Target State

AI code generators suffer from pervasive comment slop: narrating obvious syntax line-by-line (// increment counter, // return result, // check if null). Naive counters (e.g. max-added-comments: 10) break on large, well-documented pull requests while letting micro-PRs with 9 trivial comments slip through.

The target state is a metric-driven Comment-to-Code Ratio (CCR) engine that:

  1. Calculates $\text{CCR} = \frac{\text{added_comments}}{\max(1, \text{added_code})}$.
  2. Enforces an activation floor ($\ge 20$ added comments) before ratio warnings trigger.
  3. Automatically exempts markdown docs, configuration files, SQL migrations, and SPDX license blocks.
  4. Detects and blocks syntax narration slop while strictly safeguarding // SAFETY: contracts, /// Rust doc comments, and /** ... */ docstrings.

Invariant Rules

  1. Silence over noise: If a line of code is obvious to a competent engineer, narrating it with comments is banned.
  2. Contract preservation: Never strip or discourage // SAFETY: proof annotations, formal docstrings, or license headers.
  3. Proportional budget: High CCR on its own is classified as review; it is upgraded to blocker only when confirmed syntax play-by-play narration patterns are detected.

CLI Usage

# Check git diff against base commit
bunx anti-slop-engineering comment-guard --base HEAD

# Audit unified diff patch file directly with custom threshold
bunx anti-slop-engineering comment-guard \
  --diff-file diff.patch \
  --threshold 0.35 \
  --min-floor 20 \
  --json

Download now (ZIP)

Does 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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.