ravidsrkravidsrk/agent-skills/CLAUDE.md

agent-skills

CLAUDE.md1 starsChanged 21 days agoMITView on GitHub

Production-grade skills for AI agents — DNS migration, AWS migration, deep research, product finishing, viral image generation. The autonomous Orca fleets (spec-to-ship, clean-sweep, and the rest) moved to their own repo, ravidsrk/orca-fleet. This repo is capability skills only. All twelve are cataloged in README.md and mapped to trigger intents in AGENTS.md. - Every skill lives in skills/<name>/SKILL.md - YAML frontmatter requires name + description (1-1024 chars, includes both what and when) - Each skill has its own README.md for human…

In this repository13 other files it gives its agents.
- AGENTS.mdAGENTS.md
- cloudflare-dnsSkill
- deep-researchSkill
- finish-lineSkill
- fly-to-aws-migrationSkill
- knowledge-captureSkill
- knowledge-maintenanceSkill
- knowledge-querySkill
- landing-page-craftSkill
- namecheap-dnsSkill
- parallel-build-integratorSkill
- terminal-posterSkill
- visual-regressionSkill

The file

We split this CLAUDE.md into the rules it gives an agent

5 rules. About 0 are common, 2 are distinctive, and 3 are very specific to this project.

≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.

Common
Distinctive
Very specific

Distinctive:
All twelve are cataloged in README.md and mapped to trigger intents in AGENTS.md.
line 37
Very specific:
Every skill lives in skills/<name>/SKILL.md
line 42
Very specific:
Commit as the maintainer (git config user.name "Ravindra Kumar", git config user.email ravindrakumar8088@gmail.com), with no AI co-author, …
line 47
Very specific:
Always: Read SKILL.md before invoking a skill.
line 65
Distinctive:
Never: Add skills that are vague advice — every skill must be an actionable, verifiable workflow.
line 67

≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.md 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

CLAUDE.md

agent-skills

Production-grade skills for AI agents — DNS migration, AWS migration, deep research, product finishing, viral image generation.

Project Structure

skills/        → 12 capability skills (each with SKILL.md + README.md; scripts/ references/
                 templates/ assets/ as needed)
docs/          → Per-runtime setup guides + skill-anatomy spec
scripts/       → Repo-level helpers (validate-skills.py) + banner/ (repo hero reproducer)
assets/        → Banner image, screenshots
.claude-plugin/plugin.json → Claude Code marketplace manifest
LICENSE        → MIT

The autonomous Orca fleets (spec-to-ship, clean-sweep, and the rest) moved to their own repo, ravidsrk/orca-fleet. This repo is capability skills only.

Skills

Skill	Category	What it does
`cloudflare-dns`	🌐 Infrastructure	DNS migration to Cloudflare + zone hardening
`namecheap-dns`	🌐 Infrastructure	Namecheap XML API wrapper (handles wholesale-replace quirk)
`fly-to-aws-migration`	🌐 Infrastructure	7-phase Fly.io → AWS playbook with ≤9 min downtime
`deep-research`	🔍 Research	8-source parallel evidence orchestrator (X, Reddit, HN, GitHub, Polymarket, YouTube, Exa)
`finish-line`	🛠️ Dev workflow	Drives a half-built product to verified launch, fully autonomously (frozen ledger, visual checks, verifier, tamper guards, self-recovering runner)
`parallel-build-integrator`	🛠️ Dev workflow	Backlog → parallel owner agents → one gated integration branch → merged PRs, issues closed
`visual-regression`	🛠️ Dev workflow	Screenshot baselines, perceptual diffs, redaction, Playwright assertions, CI gate
`knowledge-capture`	🧠 Knowledge	One-command capture into a markdown knowledge base
`knowledge-query`	🧠 Knowledge	Grounded answers with citations, gaps, and briefings
`knowledge-maintenance`	🧠 Knowledge	Audit and repair frontmatter, links, orphans, stale pages
`landing-page-craft`	🎯 Marketing	Landing-page slop, copy, DESIGN.md, and AEO audit
`terminal-poster`	🎨 Creative	Retro-cyberpunk image posters (5 reusable templates, Nano Banana Pro)

All twelve are cataloged in README.md and mapped to trigger intents in AGENTS.md.

Conventions

Every skill lives in skills/<name>/SKILL.md
YAML frontmatter requires name + description (1-1024 chars, includes both what and when)
Each skill has its own README.md for human install + usage docs (separate from SKILL.md which is for agents)
Scripts go in skills/<name>/scripts/, references in references/, templates in templates/
Validate with python3 scripts/validate-skills.py before committing
Commit as the maintainer (git config user.name "Ravindra Kumar", git config user.email ravindrakumar8088@gmail.com), with no AI co-author, session or "Generated with" lines in commits or PR descriptions

Commands

# Validate all skills
python3 scripts/validate-skills.py

# Browse a skill
cat skills/cloudflare-dns/SKILL.md

# Install one skill into Claude Code (link, don't copy)
ln -s "$(pwd)/skills/cloudflare-dns" ~/.claude/skills/cloudflare-dns

Boundaries

Always: Read SKILL.md before invoking a skill. Set required env vars in-memory only.
Never: Hardcode secrets in scripts or commit them to git.
Never: Add skills that are vague advice — every skill must be an actionable, verifiable workflow.
Never: Duplicate skill content in the top-level README — link to SKILL.md and README.md instead.

agent-skills

The file

5 rules. About 0 are common, 2 are distinctive, and 3 are very specific to this project.

agent-skills

Project Structure

Skills

Conventions

Commands

Boundaries

See Also

Does it work?

No reviews yet

Your trail

Now do this for your own.