Skills Showdown — October 19, 2025

1

Tq (Grade A)

Security-tested development skill for Claude AI. Grade A. Troca Quente (MVP local) — restart do AppServer Protheus + healthcheck HTTP. Use quando precisar restartar o AppServer após `compile` e espera

No reviews yetFree

Full review

Troca Quente (MVP local) is a security-tested development skill for Claude AI. It restarts the AppServer Protheus and performs a health check via HTTP. This skill is used when needing to restart the AppServer after a compile and waiting for it to return for testing. It is designed for local MVP testing and does not make versioning, RPO edits, or rollback, which are features typically required for production environments. For prod-grade environments, users are directed to write their own restart commands. The skill can be encased with compile with the usage of /plugadvpl:deploy. The prerequisite for the server is for it to be registered with a restart command configured. When using the skill, the --use-server argument is required along with optional arguments such as --timeout, --no-healthcheck, --dry-run, --port, and --confirm-prod. The typical execution flow is to compile the code for various environments with the --all-envs argument and then restart the server with this skill. Common errors include the mandatory --use-server argument not being passed and the failure of the restart command due to non-zero shell commands.

• Restarts AppServer Protheus
• Performs HTTP healthcheck
• Configurable timeout and port
• Dry-run mode for testing

2

Bump Cabal Index State (Grade A)

Security-tested data-ai skill for Claude AI. Grade A. Update Cabal project index-state to the latest timestamp. Use when it needs to bump the cabal index-state.

No reviews yetFree

Full review

Bump Cabal Index State (Grade A) is a script designed to update the index-state of a Cabal project to the latest available timestamp. It automates the process of updating the index-state and generates a pull request on GitHub, if necessary. The script handles different build systems, including Nix, and is particularly useful when using haskell.nix. It also accounts for the potential delay in haskell.nix's recognition of the latest index-state compared to Cabal's. The process involves creating a temporary branch, running the cabal update command to retrieve the latest index-state, updating the index-state in the cabal.project file or other relevant files, committing the changes, and finally creating a pull request on GitHub. This script is intended for use with Claude AI and has been security-tested.

• Update Cabal index-state
• Create a new branch for changes
• Verify project build with Nix or Cabal
• Commit changes with a specific message format
• Create a pull request on GitHub

3

Document Review (Grade A)

Security-tested development skill for Claude AI. Grade A. Review requirements or plan documents using parallel persona agents that surface role-specific issues. Use when a requirements document or pla

No reviews yetFree

Full review

Document Review (Grade A) is a security-tested development skill for Claude AI that reviews requirements or plan documents using parallel persona agents to surface role-specific issues. It helps users improve their documents by dispatching specialized reviewer agents in parallel, auto-fixing quality issues, and presenting strategic questions for user decision. The skill operates in two modes: interactive and headless. In interactive mode, it engages with the user to refine and complete the review. In headless mode, it returns findings as structured text for the caller to handle, without interactive prompts.

• Parallel persona agents for role-specific issue detection
• Document type classification (requirements or plan)
• Conditional persona activation based on document content
• Auto-fix of quality issues
• Strategic question presentation for user decision

Terraform Patterns is a security-tested DevOps skill for Claude AI, specifically designed to provide reusable Azure Terraform patterns. The patterns included are utility skills that can be applied to various infrastructure setups, such as hub-spoke architecture, private endpoints, diagnostics, and AVM-TF modules. These patterns aim to simplify the process of setting up and managing Azure infrastructure using Terraform, providing a standardized approach to common infrastructure designs. The skill is categorized as a Grade A utility skill, indicating its high level of security testing and reliability. The patterns are designed to be easily integrated into existing Terraform workflows, allowing users to leverage the benefits of Azure's infrastructure capabilities while maintaining a high level of security and efficiency.

• Reusable hub-spoke Terraform patterns
• Private endpoint configurations
• Diagnostics setup
• AVM-TF module integration

5

Commit Message Generator (Grade A)

Security-tested development skill for Claude AI. Grade A. Generate structured Conventional Commit messages from staged changes, diffs, or descriptions. Use when the user asks to write a commit message

No reviews yetFree

Full review

The Commit Message Generator is a development skill for Claude AI that generates structured Conventional Commit messages from staged changes, diffs, or descriptions. It is used when a user asks to write a commit message. The skill analyzes code changes and produces well-structured Conventional Commit messages following project conventions. It works by gathering change context, classifying the change type, determining the scope, writing a subject line and body, and adding a footer if needed. The generator ensures that commit messages are in the Conventional Commits format, do not include secrets or internal information, and are concise and descriptive.

• Generates Conventional Commit messages
• Analyzes code changes and diffs
• Classifies change types (feat, fix, docs, style, refactor, test, chore, perf, ci)
• Determines scope and writes subject line and body
• Adds footer with breaking changes, issue references, or co-authors

This skill, Comparative Medicine Search, is used to find relevant human medical research that can inform veterinary clinical decisions when veterinary-specific evidence is lacking. It structures the process of searching for human evidence, assessing its applicability to veterinary patients, and clearly communicating the limitations of cross-species extrapolation. It can be applied in various scenarios such as when no species-specific RCTs or guidelines exist, or when a novel treatment being considered in animals has an evidence base in human medicine. The skill first searches for species-specific veterinary evidence, and if insufficient, it searches human PubMed for the clinical topic. It assesses the biological plausibility of cross-species application by considering factors such as pathophysiology, drug targets, and pharmacokinetic parameters. Finally, it grades the comparative evidence as Level IV and clearly labels all cross-species extrapolations in the output. This skill has been successful in various medical areas, including cardiac disease, oncology, pain management, and diabetes, but it also has limitations, such as when it fails in drug metabolism due to species-specific differences, or in behavioral medicine due to differences in cognitive and behavioral processes across species. When using comparative evidence, the skill includes a framing statement to highlight the limitations of the evidence and the need for further validation studies. Overall, the Comparative Medicine Search skill provides a valuable tool for veterinarians and researchers to inform their clinical decisions when veterinary-specific evidence is lacking.

• Searches for species-specific veterinary evidence
• Searches human PubMed for the clinical topic
• Assesses biological plausibility of cross-species application
• Grades the comparative evidence as Level IV
• Clearly labels all cross-species extrapolations in the output