I've been looking through skills in the infrastructure space, and I noticed your senior-devops skill takes on some pretty heavy lifting—orchestrating deployments, handling multi-environment configs, that kind of thing. The score suggests there's real potential here, but it looks like the spec and structure might need some refinement to really land for users.
Links:
The TL;DR
You're at 36/100, which puts this in F territory. This is based on Anthropic's Progressive Disclosure Architecture rubric for agentic skills. Your strongest area is Spec Compliance (12/15)—the frontmatter is valid and properly formatted. But Utility (2/20) is where things fall apart. The scripts and reference guides are basically placeholder templates with no actual DevOps functionality baked in.
What's Working Well
- Spec compliance is solid - Your YAML frontmatter is valid, naming conventions are correct (hyphen-case), and you've got the required metadata fields in place.
- Breadth of trigger coverage - You've listed relevant DevOps triggers like "setting up pipelines," "infrastructure as code," and "deploying applications"—good instinct on what users would search for.
- Structure exists - You've got the reference file architecture (IaC guide, CI/CD pipeline guide, deployment strategies), which shows you thought about Progressive Disclosure.
The Big One: Non-Functional Scripts
Here's the thing—all three Python scripts in scripts/ are identical placeholder templates. They don't actually generate pipelines, scaffold Terraform, or manage deployments. They just return empty self.results['findings'] = [].
This is a massive credibility killer. When someone invokes this skill expecting DevOps automation, they get stub code. This alone tanks your Utility score from 20 down to 2.
Fix: Make the scripts actually do something:
pipeline_generator.py should generate GitHub Actions workflows or GitLab CI YAML configs based on detected tech stackterraform_scaffolder.py should output real Terraform module templates (VPC, RDS, EC2 patterns)deployment_manager.py should handle orchestration logic (blue-green validation, rollback triggers, health checks)
Even basic implementations here would add 8+ points.
Other Things Worth Fixing
-
Empty reference guides - Your three .md reference files are boilerplate with generic "Pattern 1", "Pattern 2" placeholders. Replace these with actual CI/CD patterns, IaC examples, and deployment strategies. Real GitLab CI stages, GitHub Actions workflows, or Terraform modules would make this useful.
-
Missing DevOps specifics in Quick Start - You show script commands with no context. Add actual usage examples: python scripts/pipeline_generator.py ./repo --platform github --language python so users know what they're getting.
-
Generic tech stack padding - The "Supported Technologies" section (lines 106-113) lists React, Kotlin, Swift—stuff that has nothing to do with DevOps. Trim this to infrastructure-relevant stacks: Terraform, Kubernetes, Docker, AWS, GCP, etc.
-
No validation workflows - Your Development Workflow section has setup commands but no verification steps. Add: "Run npm run test to validate before deployment" and include a --dry-run flag in scripts so users can preview changes safely.
Quick Wins
- Implement actual script logic (+8 points) - Make one of the three scripts functional with real DevOps automation
- Replace reference boilerplate with real examples (+10 points) - Add concrete CI/CD configs, Terraform patterns, deployment strategies
- Add context to Quick Start (+5 points) - Show realistic command invocations with actual use cases
- Trim tech stack to DevOps-relevant tools (+3 points) - Remove frontend frameworks, focus on infrastructure
These four changes could push you into the 55-65 range pretty easily.
Checkout your skill here: SkillzWave.ai | SpillWave We have an agentic skill installer that install skills in 14+ coding agent platforms. Check out this guide on how to improve your agentic skills.
I've been looking through skills in the infrastructure space, and I noticed your senior-devops skill takes on some pretty heavy lifting—orchestrating deployments, handling multi-environment configs, that kind of thing. The score suggests there's real potential here, but it looks like the spec and structure might need some refinement to really land for users.
Links:
The TL;DR
You're at 36/100, which puts this in F territory. This is based on Anthropic's Progressive Disclosure Architecture rubric for agentic skills. Your strongest area is Spec Compliance (12/15)—the frontmatter is valid and properly formatted. But Utility (2/20) is where things fall apart. The scripts and reference guides are basically placeholder templates with no actual DevOps functionality baked in.
What's Working Well
The Big One: Non-Functional Scripts
Here's the thing—all three Python scripts in
scripts/are identical placeholder templates. They don't actually generate pipelines, scaffold Terraform, or manage deployments. They just return emptyself.results['findings'] = [].This is a massive credibility killer. When someone invokes this skill expecting DevOps automation, they get stub code. This alone tanks your Utility score from 20 down to 2.
Fix: Make the scripts actually do something:
pipeline_generator.pyshould generate GitHub Actions workflows or GitLab CI YAML configs based on detected tech stackterraform_scaffolder.pyshould output real Terraform module templates (VPC, RDS, EC2 patterns)deployment_manager.pyshould handle orchestration logic (blue-green validation, rollback triggers, health checks)Even basic implementations here would add 8+ points.
Other Things Worth Fixing
Empty reference guides - Your three
.mdreference files are boilerplate with generic "Pattern 1", "Pattern 2" placeholders. Replace these with actual CI/CD patterns, IaC examples, and deployment strategies. Real GitLab CI stages, GitHub Actions workflows, or Terraform modules would make this useful.Missing DevOps specifics in Quick Start - You show script commands with no context. Add actual usage examples:
python scripts/pipeline_generator.py ./repo --platform github --language pythonso users know what they're getting.Generic tech stack padding - The "Supported Technologies" section (lines 106-113) lists React, Kotlin, Swift—stuff that has nothing to do with DevOps. Trim this to infrastructure-relevant stacks: Terraform, Kubernetes, Docker, AWS, GCP, etc.
No validation workflows - Your Development Workflow section has setup commands but no verification steps. Add: "Run
npm run testto validate before deployment" and include a--dry-runflag in scripts so users can preview changes safely.Quick Wins
These four changes could push you into the 55-65 range pretty easily.
Checkout your skill here: SkillzWave.ai | SpillWave We have an agentic skill installer that install skills in 14+ coding agent platforms. Check out this guide on how to improve your agentic skills.