Paul Tyng
Community@paultyng · Baltimore, MD
Agent Skills by Paul Tyng
Showing 36 vetted skills indexed across 1 GitHub repositories.
audit-accomplishments
Mine cross-source work evidence into cited achievement cards for reviews.
create-decision-document
Generate standardized decision records with structured option analysis in Notion.
review-observability
Audit logging, metrics, and tracing implementations for observability gaps.
review-plan
Evaluate implementation plans and issue READY, NEEDS_REVISION, or BLOCKED verdicts with prioritized findings.
review-code
Review Go codebases and protobuf APIs across architecture, best practices, and compliance.
active-review
Generate prioritized PR review summaries and GitHub deep-linked comment drafts.
bisect
Automate git bisect to isolate the first regression-introducing commit.
push-and-watch
Push local commits, create draft pull requests, and monitor GitHub Actions CI runs.
review-api-compat
Detect breaking changes in protobuf, OpenAPI, and GraphQL API contracts.
set-session-context
Configure agent session names and colors from git branch and pull request context.
evaluate-dependency
Evaluate third-party dependencies against 10 standardized criteria and output GO, CAUTION, or NO-GO verdicts.
review-coverage
Identify untested new and modified Go code using built-in coverage tooling.
review-ci
Audit CI/CD pipeline configurations for security and supply chain risks.
analyze-knowledge
Analyze git commit history to recommend pull request reviewers and assess knowledge concentration risk.
review-infrastructure
Review infrastructure-as-code files for security, reliability, and best practice gaps.
review-documentation
Identifies gaps, inaccuracies, and drift in Go, protobuf, OpenAPI, and markdown documentation.
create-jira-item
Creates Jira stories, tasks, epics, and initiatives with automatic required custom field handling.
active-read
Run interactive active-recall study sessions over documents with source-cited evaluation.
review-database
Review PostgreSQL and MySQL migrations, queries, and schema designs.
create-skill
Create structured Cursor agent skills with CSO-optimized descriptions and validation testing.
create-rule
Create valid Claude Code rule files with YAML frontmatter and path scoping.
review-performance
Audits Go code performance risks including allocation hotspots and benchmark coverage across pull requests and full codebases.
verify-when-complete
Run fresh format, lint, build, and test verification for Go projects.
match-voice
Rewrite text to match a user's blunt, dry communication style.