candid-validate-standards_skill

This skill validates Technical.md to identify vague, overlapping, or unverifiable rules and suggests actionable improvements for clearer standards.
  • TypeScript

13

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill ron-myers/candid --skill candid-validate-standards

  • SKILL.md7.9 KB

Overview

This skill validates a repository's Technical.md to find vague, ineffective, or redundant rules. It inspects rule wording, detects overlaps with configured linters, and reports missing thresholds or unverifiable guidance. The goal is a Practical, enforceable Technical.md that adds value beyond automated tooling.

How this skill works

The skill locates Technical.md (path argument, ./Technical.md, or ./.candid/Technical.md), reads and parses rules from lists and sections while skipping code blocks and comments, and collects each rule with its line number and heading. It detects present linters (ESLint/Prettier, Flake8/Ruff, golangci-lint, RuboCop) and validates each rule for vague language, missing numeric thresholds, linter overlap, multiple concerns, and unverifiability. It outputs an organized report with severity groups and optional suggested rewrites when --fix is provided.

When to use it

  • When you want a Technical.md that is actionable and enforceable
  • Before adopting repository-wide code standards or policy changes
  • When reviews flag rules that seem subjective or redundant
  • To audit a standards file after adding new linter tooling
  • As part of CI checks to keep guidance precise and verifiable

Best practices

  • Prefer concrete criteria (e.g., "functions under 50 lines") over subjective terms
  • Remove formatting/style rules already enforced by linters
  • Split multi-concern rules into separate, single-purpose statements
  • Add numeric thresholds for quantitative guidance (limits, timeouts, sizes)
  • Make rules verifiable by code or testing; avoid guidance that requires subjective judgement

Example use cases

  • Scan a repo to detect rules that conflict with ESLint/Prettier and remove duplication
  • Refine vague team standards into specific, automatable requirements
  • Generate a prioritized report showing critical vague or unverifiable rules
  • Produce suggested rewrites for maintainers with --fix to speed adoption
  • Validate Technical.md before merging a contributing guide or policy change

FAQ

It checks for common configs: ESLint/Prettier/Biome for JS/TS, Flake8/Ruff for Python, golangci-lint for Go, and RuboCop for Ruby.

Will it modify my file automatically?

Only if you run with the --fix flag; otherwise it produces suggested rewrites and a report for maintainers to apply.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational