Warp Common Skills - Setup Guide¶

Source: warpdotdev/common-skills (411.2K combined installs) GitHub: github.com/warpdotdev/common-skills Category: Development Workflow / Spec-Driven Development First Seen: August 12, 2026 Quality Tier: 🟢 Production (Warp team, uniform ~20K installs per skill)

Warp (the terminal company) publishes the skills their own agents use daily. The cluster centers on spec-driven development: write product and tech specs, implement against them, then verify the implementation matches the spec. Around that core sit PR mechanics (review-pr, create-pr, pr-walkthrough), failure handling (fix-errors, diagnose-ci-failures, reproduce-bug-report), and meta-workflows (council, saga, research).

Installation¶

# Install the full repo
npx skills add warpdotdev/common-skills

# Or install individually
npx skills add warpdotdev/common-skills --skill review-pr
npx skills add warpdotdev/common-skills --skill write-product-spec
npx skills add warpdotdev/common-skills --skill implement-specs
npx skills add warpdotdev/common-skills --skill fix-errors

Core Skills¶

Skill	Installs	Use For
`review-pr`	20.3K	Thorough PR review
`spec-driven-implementation`	20.3K	Implement from specs
`write-product-spec`	20.2K	Product requirements docs
`write-tech-spec`	20.1K	Technical design docs
`resolve-merge-conflicts`	20.1K	Conflict resolution
`fix-errors`	20.1K	Error triage and fixes
`implement-specs`	20.1K	Turn specs into code
`create-pr`	20.0K	PR creation
`update-skill`	20.0K	Self-updating skills
`diagnose-ci-failures`	20.0K	CI failure root-causing
`council`	19.0K	Multi-perspective review council
`check-impl-against-spec`	18.6K	Verify implementation matches spec
`reproduce-bug-report`	18.4K	Bug reproduction
`validate-changes-match-specs`	16.7K	Pre-merge spec validation
`research`	13.4K	Structured research pass
`cross-critique`	13.4K	Cross-skill critique
`saga`	12.3K	Long-running task orchestration
`write-feature-docs`	10.8K	Feature documentation

Also included: brandalf, pr-walkthrough, respond-to-pr-comments-in-blocklist, scan-new-specs, readout, complain, suggestion-box.

Prerequisites¶

Requirement	Details
Node.js + npx	For the skills.sh CLI install path
Git + GitHub workflow	PR-centric skills assume GitHub
Agent runtime	Hermes/Claude Code-style agent with shell access

CorpusIQ Use Cases¶

Use Case	How
Docs & repo quality	`review-pr` + `check-impl-against-spec` on every agent-driven commit
Feature planning	`write-product-spec` before agent build tasks
CI health	`diagnose-ci-failures` for the docs build and MCP server pipelines
Long-running sweeps	`saga` to orchestrate multi-step cron workflows

Integration with CorpusIQ¶

CorpusIQ is the connector layer that lets AI assistants read the systems a business already runs on - accounting, payments, CRM, email, files and analytics - and answer with the source named. It is read-only by design: it retrieves data and never writes to, or acts in, a connected system, which is what makes it safe to point at live accounts.

This page is part of the CorpusIQ directory, the reference we maintain for the MCP server and agent-skill ecosystem. Run alongside CorpusIQ, a third-party server like this one can be combined with a company's own business data in a single answer.

See what CorpusIQ connects

Limitations / Verification¶

GitHub-centric - non-GitHub forges get partial coverage
Verify install: npx skills list | grep -E "review-pr|spec"

Powered by CorpusIQ