my-agent-skills

Plan a huge chunk of work — more than one agent session can hold — as a shared map of decision tickets on your issue tracker, and resolve them one at a time until the way is clear. On ticket close, may offer ADRs (promotion into docs/adr/) when ADR-POLICY criteria hold.

Sharpen a plan or design against this codebase via grilling + domain-modeling (CONTEXT.md / ADR offers). May offer /wayfinder when fog appears. Invoke with `/grill-with-docs`.

Applies Dockerfile, Docker Compose, BuildKit, and container security patterns for local development and hardened deployable images. Use when authoring Dockerfiles or compose files, wiring secrets (runtime vs build-time), reproducible bases, PID 1 and healthchecks, volumes and networking, supply-chain hygiene, or troubleshooting compose stacks.

Routes Go work to the right reference guides and conventions for the task. Use when working on Go architecture, implementation, refactoring, concurrency, error handling, testing, performance, or code review.

PostgreSQL database patterns for query optimization, schema design, indexing, and security. Use when writing SQL or migrations, designing schemas, troubleshooting slow queries, implementing Row Level Security, tuning connection pooling, or reviewing database changes. Based on Supabase-leaning Postgres practice.

Applies Python idioms, PEP 8 norms, typing, packaging, concurrency, tooling, and performance discipline to everyday Python code. Use when writing, reviewing, refactoring, or packaging Python; when the user names type hints, dataclasses, pytest, ruff/black/mypy, asyncio, concurrency, EAFP vs LBYL, or Python idioms; or asks for a Python-specific code-quality pass.

Implement a piece of work based on a spec or set of tickets.

Optional per-repo bootstrap of `docs/agents/issue-tracker.md` and `docs/agents/domain.md` so engineering skills know where issues live (local markdown under `work/` by default; GitHub and GitLab also supported) and how to consume domain docs. Not required for the default path — skills assume local `work/` and single-context `CONTEXT.md` / `docs/adr/` when these files are missing. Run when switching to GitHub/GitLab/custom tracker, changing tracker, or confirming multi-context domain layout.

Publish a one-session implementation plan for /implement-plan — escape hatch when the build fits one context window.

Turn the current conversation into a published spec, then validate it with /council and /best-practices-research — no interview, no tickets.

Break a plan, spec, or the current conversation into a set of tracer-bullet tickets, each declaring its blocking edges, published to the configured tracker — edges as text in one file per ticket locally, or native blocking links on a real tracker.

Diff/PR review: council → thermos/yagni → delta BPR → Sol merge. Report only.

YAGNI. Use when the user wants a lazy cut, pushes back on over-engineering, or another skill needs a YAGNI review pass. Skip non-coding requests.

Heavy one-shot implement of an attached plan (council → best practices → Karpathy → YAGNI).

Interview the user relentlessly about a plan or design. Use when the user wants to stress-test a plan before building, or uses any 'grill' trigger phrases.

Build a throwaway prototype to flesh out a design — a shareable HTML demo for state/business-logic questions, or several radically different UI variations toggleable from one route.

Captures architectural decisions as lightweight ADRs with detailed rationale, optional expanded sections, scaffolding, and index maintenance. Detects decision moments, drafts for approval, writes files, and updates docs/adr/README.md. Use when recording why the codebase chose a pattern, library, or trade-off; when the user says ADR, record this decision, or why did we choose X.

Investigate a question against high-trust primary sources and capture the findings as a Markdown file in the repo. Use when the user wants a topic researched, docs or API facts gathered, or reading legwork delegated to a background agent — including when /wayfinder fires research tickets.

Writes a compact handoff to docs/handoffs/CURRENT.md from the current Agent chat so a new session or another agent can continue without re-deriving context. Use when the user asks for a handoff, session summary, continuation doc, context for the next chat, or ending work with follow-on tasks. Triggers: "handoff", "hand off", "summary for next session", "new agent", "pick up later", "/handoff".

Runs recon on current best practices for a language, framework, or domain via live web search before implementing. Use when starting non-trivial work in an unfamiliar or fast-moving domain, library, or framework, or when the user asks to follow best practices, current practices, or "the modern way" to do something. Skip for routine edits or familiar, stable code.

Explores the codebase in a given area, spawns parallel Task subagents on Cursor first-party models for deep dives including off-angle coverage, then synthesizes results for the user's request or plan.

Architecture review: scan a codebase for deepening opportunities, present them as a visual HTML report, then grill through the candidate the user picks. Use for architecture review, deepening refactors, testability improvements, or `/improve-codebase-architecture`.

Based on Recursive Language Models (RLM) research (Zhang, Kraska, Khattab, 2025), this skill guides handling tasks that exceed comfortable context limits through programmatic decomposition and recursive sub-inquiry. Triggers on phrases like "analyze all files", "process this large document", "aggregate information from", "search across the codebase", or work spanning 10+ files or 50k+ tokens.

Pit-crew response mode for coding and agent work. Action-first, numbered steps, state restated every turn, compressed grammar by default.

Analyze an unfamiliar codebase and produce a structured onboarding guide with architecture map, conventions, and a starter AGENTS.md.

Distinctive, intentional visual design when building new UI or reshaping an existing one.

Create animation-rich HTML presentations from scratch or by converting PowerPoint files.

Awwwards-level GSAP marketing/landing page design with strict editorial typography and motion discipline.

Strict coding discipline — think before coding, simplify, surgical changes only, verifiable success criteria.

Diagnosis loop for hard bugs and performance regressions. Use when the user says "diagnose"/"debug this", or reports something broken/throwing/failing/slow.

Build and sharpen a project's domain model. Use when the user wants to pin down domain terminology or a ubiquitous language, or when another skill needs to maintain the domain model.

Explains code in a short, scannable structure with a TL;DR, sectioned ideas, and small code examples. Use when the user wants code explained, summarized, walkthroughs, or documentation-style breakdowns.

Test-driven development with red-green-refactor loop. Use when user wants to build features or fix bugs using TDD, mentions "red-green-refactor", wants integration tests, or asks for test-first development.

Teaches the user a topic over multiple sessions using a grounded learning workspace under docs/learning/, with Exa-verified resources. Use when the user invokes /teach, asks to learn a concept, wants structured lessons, or needs curated resources, glossary terms, learning records, exercises, and HTML lessons.

fixture alpha

fixture beta

Mini fixture skill alpha

Mini fixture skill alpha

Mini fixture skill beta

Creates or updates durable project documentation—README sections, architecture docs, API docs, examples, migration notes, devops docs, runbooks, and usage guides verified against code. Use when the user asks to document behavior, write or update repo docs, refresh README, add API or runbook docs, capture how something works in the repository, or remove stale documentation after code changes. Triggers: "document this", "update the README", "write a runbook", "add docs for", "document the API", "capture how X works in the repo", "/document".