modiqoOfficial·22 Agent Skills Included

skillspec

Turn prose AI skills into enforceable, auditable execution contracts

Audits any SKILL.md for buried rules, undeclared tools, and missing proof, then converts it into a structured contract your agent must follow. Eliminates silent rule-skipping, unverifiable 'done' claims, and bloated prompt files that agents ignore. Compiles, tests, installs, and traces skills across Claude Code, Codex, and other harnesses with an inspectable run record.

820|59

↗

npx skills add modiqo/skillspec --all -g -y

All Skills in This Repository (22)

Pure Emerald Level Indicators

📦 In Repo

modiqo

skillspec

Inspect skill repositories and import SKILL.md skills from local folders or public URIs.

rote-shell

Execute commands, capture output, follow logs, and crystallize CLI work into TypeScript flows.

generic-research-codebase

Generate structured architecture reports from codebases using ripgrep and Git.

generic-commit-convention

Validate conventional commit messages and run pre-push checks.

generic-code-review

Automate code review by interpreting skill specification files.

rote-browse

Automate browser tasks with rote commands and page leases.

research-codebase

Analyze codebase structure and components using parallel sub-agents.

commit-convention

Validate commit messages and PR titles against the Conventional Commits standard.

code-review

Orchestrate parallel code reviews and synthesize findings into a structured report.

browser-research-prose

Automates web browsing, inspection, snapshotting, and evidence extraction from pages.

demo-browser-research

Automate web-based research tasks with browser automation and evidence logging.

dev-repo-readiness

Compare local repository state against remote repositories and CI configurations.

Official

Advanced

Frequently Asked Questions

FAQPage Schema

How to install SkillSpec?▼

Run `npx skills add modiqo/skillspec --all -g -y` in your terminal to install the full suite globally.

What problem does SkillSpec solve?▼

It checks whether your AI agent actually followed a skill's rules, then turns the skill into a structured contract with routes, forbidden actions, and proof requirements so nothing gets silently skipped.

How do I check if a skill is risky before installing it?▼

Run `skillspec doctor` on any local folder or public GitHub URL to get a risk report covering buried rules, undeclared dependencies, and missing proof, or use the hosted page at skillspec.sh.

Does SkillSpec work with Claude Code and Codex?▼

Yes. It installs as a plugin for Claude Code and Codex and works with any harness that supports the SKILL.md standard, without replacing your existing agent runtime.

Do I need to rewrite my existing skills to use SkillSpec?▼

No. The import command converts your existing prose SKILL.md into a structured skill.spec.yml automatically, keeping the original file in place.