All authors
Claude Skills by tompassarelli
github.com/tompassarelli44 skills0 installs22 views
- Babashka DevelopmentDevelop Babashka scripts, tasks, runtime upgrades, and C interop using the consumer's runtime and current official capabilities.Votes: 0GitHub stars: 3
- Ceremony BudgetDecide how much hardening, security, signing, compatibility or test isolation a change needs before adding any, and harden one named guarantee when it does. Use when tempted to add process, safeguards or "while I'm here" fixes.Votes: 0GitHub stars: 3
- Competitive Development LoopMeasure and improve a named repeated edit-to-result loop when its latency materially affects iteration speed.Votes: 0GitHub stars: 3
- DebuggingDiagnose and fix a non-trivial bug, regression, crash, or unexplained behavior by researching the symptom, isolating one variable at a time, and diffing good against bad runs.Votes: 0GitHub stars: 3
- Effect DevelopmentDesign, implement, review, or migrate TypeScript systems using the Effect ecosystem. Use for Effect APIs, service composition, typed errors, schemas, resources, concurrency, streams, and Effect-specific tooling. Also load it before designing any host tool, command or runner in a repository that has Effect installed when the tool starts processes, waits, retries, holds a resource or parses outside data.Votes: 0GitHub stars: 3
- External CodeCheck licensing, attribution, and derivation rights before copying, adapting, vendoring, or using outside code as an implementation reference.Votes: 0GitHub stars: 3
- Importing SkillsImport third-party skills into the shared source catalog and activate them for the configured consumers.Votes: 0GitHub stars: 3
- PlanningPlan a change when a concrete interface, durable state, dependency, or external boundary makes the decision costly to reverse. Skip routine changes with a clear method and check.Votes: 0GitHub stars: 3
- Prior ArtInvestigate existing solutions when evidence must decide a consequential protocol, representation, dependency, interface, or infrastructure choice.Votes: 0GitHub stars: 3
- Program CraftsmanshipRefactor, clean up or port established code while preserving observable behavior; ports are written in the target language's idioms.Votes: 0GitHub stars: 3
- Program StewardshipScope delivery work, diagnose stalled completion, choose engineering investment, and consolidate issues without losing independently finishable outcomes.Votes: 0GitHub stars: 3
- Rust DevelopmentDevelop, review, debug, test, or optimize Rust and Cargo code, including ownership, errors, async, unsafe boundaries, and build behavior.Votes: 0GitHub stars: 3
- Skill MaintenanceTurn explicit lasting instructions or demonstrated recurring lessons into scoped, source-owned agent policy.Votes: 0GitHub stars: 3
- ThreejsBuild or debug Three.js scenes, covering setup, geometry, materials, textures, lighting and shadows, asset loading, animation, picking and controls, shaders and postprocessing.Votes: 0GitHub stars: 3
- VerificationChoose which proportionate evidence proves a claim and interpret its result, including delayed or ambiguous runs. Writing or pruning tests belongs to the testing skill.Votes: 0GitHub stars: 3
- Agent PolicyAuthor, locate, register, or activate source-owned agent instructions, skills, hooks, and modules.Votes: 0GitHub stars: 3
- Agent Runtime IncidentInvestigate and close unexplained agent admission, startup, death, liveness, control, or reporting failures at their owning cause.Votes: 0GitHub stars: 3
- Auto WakeAuto-wake mode: when the operator asks for "auto-wake" (or to keep working through usage limits, overnight or while away), schedule a recurring session tick that resumes the work after a pause and keeps it moving.Votes: 0GitHub stars: 3
- Automation ChromeDrive unattended web work in dedicated private headful Chrome with Playwright, screenshots, and page-to-page secret transfer.Votes: 0GitHub stars: 3
- Cloud WorkersRun every code-only worker task in Anthropic's cloud first (Proton credit lane, then RemoteTrigger routines); local workers take only Warcraft clients, private game assets, the LAN pool or unpushed local state.Votes: 0GitHub stars: 3
- ConvoRecall past agent work, decisions, commands or session IDs across Claude Code and Codex transcripts with the indexed convo CLI, instead of searching transcript storage.Votes: 0GitHub stars: 3
- Game Design PrototypingDesign and build playable game prototypes and worlds using Tom's local Quaternius models, scenery, and animation library; make navigation and combat legible without placeholder actors.Votes: 0GitHub stars: 3
- Game IpRespect Blizzard's and Nintendo's IP in any game-modding or game-feel work (Warcraft III, Wisp, Smashcraft, Melee): where engine or game facts may come from, and what to do when a task seems to need the game's internals.Votes: 0GitHub stars: 3
- Gemini ReviewGet an extra adversarial review of a public diff or design from Gemini's strongest Pro model through the Antigravity CLI (`agy`) on Tom's Google AI plan.Votes: 0GitHub stars: 3
- Github ActionsWrite or change GitHub Actions workflows, dispatch or wait on runs, or run many agents against GitHub within the account's job limit and hourly API budget: cap matrices, poll with backoff, clean up scratch branches.Votes: 0GitHub stars: 3
- Github IssuesCreate or edit GitHub issues, including issue labels, using the repository's existing ticket and tagging conventions.Votes: 0GitHub stars: 3
- Guard AuthoringImplement or change a command/write guard and its shared identity, activity gate, provider wiring, and deny/pass fixtures.Votes: 0GitHub stars: 3
- Image Context BudgetInspect repeated screenshots or graphical sessions without accumulating image payloads in conversation history; recover from image-related request-size failures.Votes: 0GitHub stars: 3
- Lua PerformanceCut frame cost, Lua instruction counts or per-frame allocation in TypeScriptToLua code on PUC Lua 5.3 (Warcraft III, Wisp): over-ceiling Lua32 tests, failing `bun wisp perf compare`, GC pressure.Votes: 0GitHub stars: 3
- Machine CapacityBound sustained multi-core or >1 GiB local work and diagnose agent-caused resource pressure. Skip ordinary edits and small checks.Votes: 0GitHub stars: 3
- Nix DevelopmentCreate or repair project-local Nix environments and consumed flake outputs outside nixos-config.Votes: 0GitHub stars: 3
- Orchestrating CodexRun work through a Codex lead session that Claude supervises: start it, give it a goal, message it, watch its worker tree and keep it on a tight leash. Use when Tom asks to drive work through Codex, or for a Codex lead or orchestrator.Votes: 0GitHub stars: 3
- Private Desktop DevelopmentRun GPU-accelerated Linux games and graphical development tools in a private desktop beside Niri, with remote input that never grabs the normal desktop.Votes: 0GitHub stars: 3
- Project StructureOrganize project files by real authority, lifecycle, and retrieval needs, including data pipelines and handoff layouts.Votes: 0GitHub stars: 3
- Repo SafetyEdit and publish ~/code repositories through owned worktrees, enumerated staging, and safe-push; preserve main checkouts, pins, and peer work; keep disc images and extracted game files out of Git trees.Votes: 0GitHub stars: 3
- Resource Safe SearchSelect bounded filesystem roots before searching repository containers or virtual filesystems; use native process metadata for process discovery.Votes: 0GitHub stars: 3
- Smashcraft AnimationAuthor and improve Smashcraft fighter animation (readable silhouettes, drills, locomotion, rolls, get-up attacks, grabs and throws, nine-way damage reactions) for quality and action coverage.Votes: 0GitHub stars: 3
- Smashcraft Character CreationDesign, build and test a Smashcraft fighter or move (role, move phases, hit regions, damage, knockback, shield, recovery, rollback state, selection) for a new fighter or move change, not other Warcraft maps.Votes: 0GitHub stars: 3
- Smashcraft Stage DesignDesign, dress and judge a Smashcraft stage's look (backdrop, scenery, set pieces, water, sky, light, fighter contrast) in Classic and Definitive, for any stage art, composition, look issue or new stage.Votes: 0GitHub stars: 3
- TestingWrite, change, delete, run or speed up automated tests in any project: designing a test away, what earns a test, property, replay and golden tests, CPU cost, flaky tests, pruning a suite, landing gates and reuse by input hash.Votes: 0GitHub stars: 3
- TodoKeep restart-grade records for cross-turn work, external waits, parked tasks, and handoffs. Also handles an explicit cross-supervisor protocol request.Votes: 0GitHub stars: 3
- Warcraft ModdingUse for any Warcraft III map work: Wisp TypeScript maps (hot reload, headless and 32-bit Lua tests), the native game (LAN and signed-in clients, startup recovery, desync debugging), and Smashcraft's ts/.Votes: 0GitHub stars: 3
- WorkersAssign independent work, choose provider tiers and ETAs, and run workers from Codex or Claude.Votes: 0GitHub stars: 3
- Wurst DevelopmentDevelop Warcraft III maps and tooling in WurstScript (projects, compiler workflows, tests, compile-time object data, WC3 UI) when the requested source or toolchain is Wurst; existing Lua or Jass projects need no migration.Votes: 0GitHub stars: 3