Andrew Codesit
Community@andrewcodesit · Prague
Andrew Codesit provides agent-driven software development lifecycle skills covering spec definition, planning, task tracking, code review, TypeScript auditing, UI verification, and git publishing.
Agent Skills by Andrew Codesit
Showing 13 vetted skills indexed across 1 GitHub repositories.
andrewcodesitinit-context-files
Creates a structured context directory with AGENTS.md files for agent project onboarding.
andrewcodesitbreakdown-tasks
Breaks specs and feature requests into small trackable task and epic files.
andrewcodesithandoff
Generates a session handoff document capturing goals, decisions, and next steps for another agent.
andrewcodesitplan
Creates grounded, phased implementation plans verified against the actual codebase.
andrewcodesitclose-task
Finalizes tasks by running validation, marking merge requests ready, and updating task status.
andrewcodesitdefine-spec
Creates concise implementation-ready specs from project ideas, feature requests, or vague goals.
andrewcodesitplan-review
Reviews implementation plans against specs, codebase reality, and design rules before execution.
andrewcodesitgit-publish
Commits, pushes, and opens a draft merge or pull request on the connected git host.
andrewcodesitcode-review
Reviews code diffs for correctness, contract safety, and structural quality with cited findings.
andrewcodesitverify-ts
Audits TypeScript code for type-safety, strictness, and idiomatic patterns against a defined rule set.
andrewcodesitverify-ui
Verify local UI changes in a real browser by exercising flows and inspecting console and network errors.
andrewcodesitcleanup
Scans repositories for dead code, debug artifacts, and unused imports, then applies confirmed fixes.
andrewcodesitexecute
Implements an approved plan and verifies the result against its contracts and repo standards.
Frequently Asked Questions About Andrew Codesit
FAQPage SchemaWhat tasks can I accomplish with Andrew Codesit's skills?▼
You can initialize project context files, define implementation specs, write and review plans, break specs into tracked tasks, execute approved plans, run strict code reviews, audit TypeScript type-safety, verify UI changes in a browser, clean dead code, and publish branches as pull requests.
Who are these skills designed for?▼
Software engineers and development teams working with agent-assisted coding sessions. They suit developers who want structured spec-to-implementation flows, TypeScript strictness enforcement, UI verification, and repo hygiene checks integrated into their daily branch and pull-request routine.
How do the skills fit into a typical development session?▼
Start with define-spec and plan, break work into tasks, then trigger execute to implement against the plan. Use code-review, verify-ts, or verify-ui for validation, close-task to finalize review readiness, and git-publish to commit, push, and open a draft merge or pull request.
Do the skills support external project-management systems?▼
Yes. The breakdown-tasks and close-task skills support local task files with optional epics, or connected external project-management systems for creating tickets, issues, and marking tasks done. Git-publish connects to your git host to open draft merge or pull requests.
What quality checks are available before merging code?▼
Three verification skills exist: code-review checks structural quality and contract safety, verify-ts enforces TypeScript strictness including no-any and branded types, and verify-ui exercises local pages in a browser checking layout, console, and network errors. Cleanup removes dead code and debug artifacts.