skills.rest
rubrical-works
rubrical-worksOFFICIAL
Explore all skills by this author →
GitHub

command-spec-audit

Evaluates command specification files for structural, decision, and execution formatting weaknesses.

Updated Mar 30, 2026
Search Tags:#spec review#llm reliability#extension points#markdown formatting#severity rubric#command audit#destructive patterns
One-click install
npx skills add https://github.com/rubrical-works/idpf-praxis-skills --skill command-spec-audit-rubrical-works
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: command-spec-audit
Source: https://github.com/rubrical-works/idpf-praxis-skills/tree/main/Skills/command-spec-audit
Command: npx skills add https://github.com/rubrical-works/idpf-praxis-skills --skill command-spec-audit-rubrical-works

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Command specifications are structured natural-language programs whose formatting acts as a parsing contract for LLMs; weak formatting degrades execution reliability, especially after context compaction. This Skill provides a shared evaluation rubric to detect and fix those weaknesses before deployment. ## Core Features & Use Cases - Five-Category Rubric: Audits structural integrity, decision formatting, execution reliability, extension points, and destructive pattern detection with severity-ranked findings. - Destructive Pattern Scanning: Flags dangerous commands in extension blocks such as rm -rf, git push --force, DROP TABLE, and --no-verify with context-aware severity levels. - Standardized Output: Produces a findings table with criterion, severity, location, and recommendation, plus a High/Medium/Low summary count. - Use Case: Before shipping a new slash command spec, run the audit to catch oversized steps, missing STOP boundaries, unfenced code blocks, and mismatched extension markers. ## Quick Start Audit my command specification file for formatting weaknesses and report findings by severity.

Frequently Asked Questions about command-spec-audit

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I audit a command specification file for quality issues?▼

Apply the five-category rubric covering structural integrity, decision formatting, execution reliability, extension points, and destructive patterns. Each finding is reported with severity, location, and a concrete fix recommendation in a standardized table.

What formatting issues affect LLM command execution reliability?▼

Key issues include steps over 30 lines without sub-headers, STOP boundaries buried in paragraphs, commands in inline backticks instead of fenced blocks, and vague conditions like "if needed". These degrade parsing, especially after context compaction.

How are destructive commands detected in extension blocks?▼

The audit scans USER-EXTENSION blocks for patterns like rm -rf, git push --force, DROP TABLE, and --no-verify, assigning Critical or Warning severity. Context rules exempt expected cases like temp file cleanup and scoped deletions.

What severity levels does the command spec audit use?▼

Findings are ranked High, Medium, or Low. High severity issues degrade LLM reliability and must be fixed before deployment, Medium issues reduce clarity after compaction, and Low issues are style improvements.

When should I run a command specification audit?▼

Run it before deploying new command specs, when revising existing commands, and as a creation-time guardrail for extensions. It serves as a shared rubric for audit and extension-creation workflows.

Related Skills

Explore More Software Engineering Skills→
alekspetrovalekspetrov
196

nav-marker

Create Markdown context markers with summaries and next steps.

Community
Basic
davidleedavidlee
26

spec-driver

Create, list, find, show, edit, and complete spec-driver entities via CLI.

Community
Intermediate
Fabric-ProFabric-Pro
3

code-review

Reviews pull requests and code changes for correctness, security, performance, and maintainability.

Official
Intermediate
compozycompozy
96

spec-peer-review

Facilitate cross-LLM peer reviews of code specifications with a specified model.

Official
Advanced