Agent skills for software development, debugging, code review, and implementation. Browse reusable SKILL.md agent workflows and prompt templates in this category. Each card links to when-to-use notes and setup steps.
Connect the foreman's session to a builder in an Orca pane, Pi or Claude Code (any Claude model), and PROVE the link (a pane the foreman opened itself (or one the owner confirmed) + assigned session id, two-round handshake, Stop-log and session-file evidence), then send work-order pointers and wait on the builder's real turn-end signal. Use for /orca-pi-connect, "connect to pi", "connect to the claude builder", "set up the builder pane", or before a Topping Out job whose repo profile says builder delivery = pane link. Orca only (another pane host: hand-carry the pointers). Proves and uses the channel; never decides or approves anything for the owner. Also answers to its old name, /pi.
PostgreSQL database architecture, optimization, and AWS Aurora expertise. Use when working on schema design, query optimization, migrations, or database operations. Do NOT use for application code or other databases.
Manual only. Invoked by the owner to cut a MailFathom release — composes the changelog, raises the declared version, settles the milestones and the next release's tracking issue, and states the order the two pull requests and the tag have to land in.
Use this skill to work the open PR queue end-to-end: assess merge readiness, run security gates, selectively request Claude Opus review on substantial PRs, merge eligible PRs, and hand off release PRs to /release.
Obfuscate / protect Python source before shipping it, and verify the obfuscated output still runs. Use when the user wants to protect, obfuscate, harden, or "make it hard to read / reverse-engineer" a Python package, CLI, or service before publishing to PyPI, sending a build to a client, or shipping an evaluation copy — especially when they want confidence the transform didn't break the code. Powered by pyobfus (an AST-based, framework-aware, AI-debuggable obfuscator). Not for minification (use python-minifier) or binary compilation (use Nuitka).
Review a Python project for source-protection readiness without changing anything: what would break under obfuscation, which framework preset fits, which files the current config already excludes, and exactly which artifacts a build would emit. Use when the user asks "is this safe to obfuscate?", "what would protecting this break?", "review before we ship", or wants a pre-merge / CI check. Read-only by design — it never obfuscates, never writes deliverables, and never runs project commands. To actually produce a protected build, use the pyobfus-protect skill instead.
Ruby on Rails development following Rails Way conventions, RSpec testing, and security best practices. Use when working on Rails projects requiring MVC architecture, Active Record, or API development. Do NOT use for other programming languages or frameworks.
Expo and React Native mobile development for iOS and Android. Use when building cross-platform mobile apps, native modules, or mobile-specific UI. Covers React Navigation, platform-specific code, and mobile performance. Do NOT use for React web development.
Transform monolithic Terraform configurations into reusable, maintainable modules following HashiCorp's module design principles and community best practices.
Code review skill for PRs, branches, or uncommitted changes. Loads the appropriate subset of architecture/subsystem docs based on what changed, walks the changeguardrailsrules pre-PR checklist, and emits structured findings (blocking / advisory / nit) with a verdict.
Rust systems programming with zero-cost abstractions, memory safety, and TDD. Use when working on Rust projects requiring performance, safety, or systems-level code. Emphasizes std-first philosophy and minimal dependencies. Do NOT use for other programming languages.
Consulta a base LOCAL de reuniões do ScribaDev (índice SQLite+FTS5 com resumos, transcrições, participantes, clientes e pendências). Use sempre que o usuário perguntar sobre reuniões ou calls gravadas, mesmo sem citar o Scriba: "qual foi a última reunião do cliente X", "o que foi discutido/decidido sobre Y", "quem participou da call de sexta", "quais pendências ficaram da reunião", "em que reunião falamos de Z", "me resume a reunião de ontem".
Make, read or edit an Excel workbook (.xlsx) - a table of numbers, a budget, a tracker, a model with formulas - with a TypeScript script run by bise office run. Use when the user asks for a spreadsheet or an xlsx, sends one, or leaves notes on an xlsx artifact.
POSIX-compliant shell scripting and automation with focus on portability, robustness, and Unix philosophy. Use when writing bash scripts, automation, or CLI tools. Includes BATS testing patterns. Do NOT use for other programming languages.
Evolutionary optimize-by-metric loop — mutate a target, evaluate against fixed criteria, KEEP if the score improves (commit) or REVERT if it doesn't (git checkout), repeat, plateau-break after N stagnated runs. Adapts Karpathy's autoresearch pattern (and the ECC autoresearch-agent) with mandatory yool guardrails (§11 caps), git-isolated branch discipline, an anti-Goodhart eval order (correctness gate FIRST, score SECOND, labeled tokenizer), a local-first mutation ladder, and a simplicio.savings-event/v1 receipt per run. Use when the user says "optimize this by <metric>", "autoresearch loop", "evolutionary optimize", "mutate/eval/keep-revert", or asks to shrink tokens/latency/bundle size/improve pass-rate against a fixed, measurable eval. Worker: scripts/autoresearch.py.
Make, read or edit a PowerPoint deck (.pptx) - a review, a pitch, a status deck, slides with tables and charts - with a TypeScript script run by bise office run. Use when the user asks for slides, a deck or a pptx, sends one, or leaves notes on a pptx artifact.
Check a TLA+ model of a system that moves between states, such as a product flow, a protocol, or a service, and draw it as a Mermaid statechart. The report gives the shortest path to each state that breaks a rule. Use for /statechart, or to show a bug as a path through states.
Repo-specific PR/CI conventions for an agent subscribed to PR activity (loopdive/js2wasm). Read this BEFORE acting on a CI or review event. Overrides the generic PR-steward defaults where this repo's merge queue, gates and hooks make them wrong — most importantly: agents NEVER enqueue, a bot hold is a real regression, and close+reopen is the sanctioned way to DEQUEUE for a push.
Suspend ALL in-flight team work durably — snapshot every agent worktree, write a ## Suspended Work handoff into each lane's issue file, push every lane branch, open one DRAFT PR per lane with the handoff in the description, and disarm the loop's timers. Run before a planned stop, container teardown, or long pause. Resume is a fresh agent per lane reading the issue file's handoff.