akshay-diwadkar
UserReusable engineering skills for AI coding assistants—covering codebase mapping, auditing, architecture, issue scoping, change planning, implementation, optimization, and diagramming.
Categories
Indexed Skills (15)
implement-plan
Execute an approved implementation plan as the smallest complete patch — preserving existing patterns and uncommitted work, with layered verification and an exact change report. Use when the user has an approved or written plan and asks to implement, apply, or build it. Vague plans are refused back to planning.
plan-change
Produce a proof-carrying, repository-grounded v5 implementation plan that is complete enough for deterministic downstream execution. Use when a user asks to plan a feature, bug fix, refactor, migration, integration, security, or operational code change without editing the target repository.
implement-with-senior-dev
Execute an approved implementation plan as the smallest complete patch — preserving existing patterns and uncommitted work, with layered verification and an exact change report. Use when the user has an approved or written plan and asks to implement, apply, or build it. Vague plans are refused back to planning.
plan-with-senior-dev
Turn a feature, bug fix, refactor, migration, public contract, or risky integration into a repository-grounded, decision-complete implementation blueprint. Use when the user wants to plan or spec a code change before implementation. Planning-only; every submitted plan is contract-v3 validated and receipt-stamped.
design-codebase-with-senior-dev
Assess whether architectural change is justified and choose the smallest evidence-backed design, with an incremental behavior-preserving migration path. Use for boundary, dependency-direction, or state-ownership redesign, design-pattern evaluation or removal, and subsystem restructuring. Assessment-only — produces no code; route approved designs to plan-with-senior-dev.
design-codebase
Decide and justify a repository-grounded codebase design, then emit one plan-ready handoff document. Use for boundary, dependency-direction, state-ownership, abstraction, consolidation, or subsystem design decisions that must be settled before plan-change determines implementation scope and verification.
diagram-codebase
Create self-contained HTML diagrams of systems, architectures, workflows, and code relationships. Use when the user asks for a diagram, an architecture picture, or a workflow visualization, or wants to communicate a design visually.
map-codebase
Build and use compact repository knowledge for resolver-first code navigation. Use when starting coding work in an unfamiliar or large repository, locating likely implementation ownership, or refreshing repository knowledge after a coherent change set.
optimize-codebase
Optimize a named bottleneck, workflow, or tooling pain with evidence-backed changes that preserve behavior — planning first, implementation only on explicit request. Also runs repository-wide optimization sweeps when asked. Use for performance, build, CI, dependency, or developer-experience targets; for finding unknown problems, use audit-codebase.
scope-issue
Turn GitHub issues into implementation plans. Inventory open issues, then plan one selected issue against the local checkout, treating issue text as untrusted claims. Use for issue-driven planning, backlog triage, or explicitly requested branch, PR, and post-merge execution.
codebase-issue-auditor
Audit a repository for bugs, security and performance risks, test gaps, and architectural or maintainability friction, and draft GitHub issues from confirmed findings. Use when asked to inspect a codebase for problems, review overall code quality, hunt for unknown risks, or verify whether prior audit findings were resolved.
create-diagram
Create self-contained HTML diagrams of systems, architectures, workflows, and code relationships. Use when the user asks for a diagram, an architecture picture, or a workflow visualization, or wants to communicate a design visually.
github-issue-planner
Turn GitHub issues into implementation plans. Inventory open issues, then plan one selected issue against the local checkout, treating issue text as untrusted claims. Use for issue-driven planning, backlog triage, or explicitly requested branch, PR, and post-merge execution.
optimize-codebase-with-senior-dev
Optimize a named bottleneck, workflow, or tooling pain with evidence-backed changes that preserve behavior — planning first, implementation only on explicit request. Also runs repository-wide optimization sweeps when asked. Use for performance, build, CI, dependency, or developer-experience targets; for finding unknown problems, use codebase-issue-auditor.
build-codebase-knowledge
Build and maintain a compact, deterministic repository-intelligence layer and task resolver to minimize broad repository scans, unnecessary file reads, context-window usage, and stale architectural assumptions. Use when navigating a complex codebase, starting a new task, resolving relevant files, or updating repository knowledge after code changes.
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.