Hmbown
- 469 skills
- 0 followers
- last month last updated
- ▌ Locate Animals Or Plants 2 · hmbownLocate Animals or Plants is service discovery with ecological instincts. It helps you find running services, background jobs, data stores, dependency roots, and other living things in an environment before you touch them. The point is not just location, but context: what is active, who depends on it, and what keeps it alive.
- ▌
- ▌ Code Clarifier · hmbown bundleClarifies and documents code for readability, intent, and comprehension while preserving all functionality. Focuses on recently modified code unless instructed otherwise. Use when the user asks to clarify code, add documentation, improve naming, explain intent, make code self-documenting, improve readability, add context, or make code easier to understand. Also use when the user says "clarify", "document this", "make this readable", "what does this do", or "explain this code by improving it".
- ▌
- ▌ Taste · hmbown bundleDomain-grounded judgment for making things — code, documents, UI, charts, plans, messages — and for critiquing them. Use when the user asks for taste, judgment, or polish ("make this good", "tighten this", "this feels off", "this feels generated"), when a deliverable is externally visible or high-stakes, when the brief is thin and the pull is to fill gaps with plausible defaults, or when asked to evaluate or compare work. Not for verbatim transformation or exhaustive extraction.
- ▌ Trimtab · hmbown bundleUse when the user asks for /trimtab or wants to bootstrap or run a closed-loop self-verifying agentic workflow in a repo. Defaults to the safest profile: Claude orchestrates and may use Claude or Codex sub-agents for work, while a fresh Codex context verifies closure.
- ▌ Init Trimtab · hmbown bundleUse when the user asks for $init-trimtab or wants to inspect a specific repo and install or retune a Trimtab player/coach workflow for that project's actual constraints, tools, and task graph.
- ▌ Arxiv Search · hmbown bundleSearch arXiv preprint repository for papers in physics, mathematics, computer science, quantitative biology, and related fields
- ▌ Query Writing · hmbownFor writing and executing SQL queries - from simple single-table queries to complex multi-table JOINs and aggregations
- ▌ Schema Exploration · hmbownFor discovering and understanding database structure, tables, columns, and relationships
- ▌ Run · hmbownUse the local Autodialectics MCP server and CLI to compile tasks, execute anti-slop runs, inspect artifacts, replay runs, benchmark policies, and evolve champions in this repository.
- ▌ Replay · hmbownRe-run a stored Autodialectics pipeline run, optionally with a different policy, to compare outcomes or debug regressions.
- ▌ Inspect · hmbownExamine Autodialectics run results, manifests, and stored artifacts. Use after a pipeline run completes or to review past runs.
- ▌ Benchmark · hmbownRun benchmark suites and manage policy evolution — create challengers, compare against champions, promote or rollback policies.
- ▌
- ▌ Photo Learning · hmbownRecognize a photo and narrate a kid-friendly explanation using image understanding + TTS.
- ▌ Short Drama Kit · hmbownGenerate a short script, character voices, and an optional teaser for a mini drama.
- ▌ Audiobook Studio · hmbownProduce a full multi-chapter audiobook with consistent narration and exportable files.
- ▌
- ▌ Social Media Kit · hmbownGenerate vertical TikTok/Reels videos with captions, trending music, and engaging hooks.
- ▌ Storybook Lesson · hmbownCreate a kid-friendly learning card with an illustration and narrated audio.
- ▌ Voiceover Studio · hmbownDesign custom voices from text prompts and produce professional narration with voice design previews.
- ▌ Ad Commercial Kit · hmbownCreate professional commercials with hook narration, product demonstration, music, and call-to-action.
- ▌ Voice Podcast Kit · hmbownCreate multi-host podcast episodes with cloned voices, intro music, and transitions.
- ▌ Audiobook Workshop · hmbownTurn long text or a file into an audiobook using async TTS, voice selection, and file tools.
- ▌
- ▌ Interactive Story Kit · hmbownCreate branching audio stories with distinct character voices and atmospheric music that shifts with narrative mood.
- ▌ Language Learning Kit · hmbownGenerate vocabulary flashcards and pronunciation guides with native audio and cultural context images.
- ▌ Music Video Generator · hmbownCreate synchronized music and matching video from a unified creative prompt.
- ▌ Educational Course Kit · hmbownGenerate complete micro-courses with lesson scripts, slide images, narration audio, and preview videos.
- ▌ Meditation Wellness Kit · hmbownGenerate calming audio-visual experiences for meditation, sleep, and relaxation.
- ▌ Product Photography Kit · hmbownTransform product photos into e-commerce ready images with backgrounds, lifestyle contexts, and marketing variations.
- ▌ Trailer Teaser Generator · hmbownCreate cinematic trailers with voiceover, tension-building music, and dramatic visual sequences.
- ▌
- ▌
- ▌ Run Cycle · hmbownChoose a Knights Radiant cycle and hand-off sequence for greenfield, risky migration, triage, refactor, or small-task work.
- ▌
- ▌
- ▌
- ▌
- ▌
- ▌
- ▌
- ▌
- ▌ Choose Order · hmbownChoose the correct Knights Radiant order for the task before starting execution.
- ▌
- ▌ PDF · hmbownRead, extract, split, merge, rotate, watermark, fill, OCR, or create PDF files with verification of page counts and text extraction.
- ▌
- ▌ Help · hmbownRoute a "how do I use Codewhale" question to the installed help, config, and doctor surfaces instead of reciting a manual from memory. Explicit-only.
- ▌ Plan · hmbownTurn a sufficiently understood task into an ordered implementation plan with dependencies and verification. Orchestrate Codewhale’s native plan state; do not build a parallel planner.
- ▌
- ▌
- ▌ Debug · hmbownReproduce, minimize, localize, identify root cause, and distinguish diagnosis from an authorized fix. Prefer root-cause over symptom patches.
- ▌ Feishu · hmbownWork with Feishu or Lark bots, docs, sheets, bitables, approval flows, and OpenAPI/MCP setup without hardcoding credentials.
- ▌ Review · hmbownDiff-scoped correctness review that reads the codebase around the change — callers, contracts, and invariants — and returns line-anchored findings ranked by severity with confidence, then a merge-risk verdict. Use for reviewing a PR, diff, or named change set. Not for style review, approvals-as-rubber-stamp, or editing the code.
- ▌ Verify · hmbownExercise the real app/API/CLI and collect observable evidence; tests alone do not count as end-to-end verification.
- ▌ Dataviz · hmbownChoose and produce an appropriate chart/dashboard/visual explanation from data, with legible encodings and source/assumption notes.
- ▌ Handoff · hmbownWrite a compact, decision-ready handoff so the next session (or the user) can continue without reconstructing the current one. Use when the session is ending, context is running low, the user asks for a handoff / 'pass the baton' / 'hand off', or a long-running operation needs a durable state checkpoint.
- ▌
- ▌
- ▌ Research · hmbownCurrent external research using available web tools, primary sources where possible, links/citations, date awareness, synthesis, and unresolved uncertainty.
- ▌ Simplify · hmbownImprove clarity and reduce needless complexity after behavior is understood; preserve behavior and keep cleanup separate from correctness fixes.
- ▌
- ▌ Implement · hmbownCarry an authorized, defined request or approved plan through scoped edits and proportionate verification. Skill loading is not write authorization by itself.
- ▌ Interview · hmbownAsk one useful structured question at a time only when material product/implementation choices are genuinely missing; remember answers and produce a brief/spec. Discoverable facts should be investigated instead of asked.
- ▌ MCP Builder · hmbownDesign, build, configure, or debug Model Context Protocol servers for codewhale, including stdio and HTTP/SSE transports.
- ▌
- ▌
- ▌ Skill Creator · hmbownCreate or improve codewhale skills. Use when the user wants a new skill, wants to update an existing skill, or needs guidance on when a skill should be a skill versus MCP, hooks, tools, or a plugin scaffold.
- ▌ Webapp Testing · hmbownStart/reuse a local app, wait for readiness, inspect rendered state/console/network, act from observed selectors, and verify with evidence.
- ▌ Frontend Design · hmbownProduce intentional, responsive, accessible UI work and perform visual QA instead of generic component assembly.
- ▌ Security Review · hmbownReview a change, module, or surface for exploitable defects — trust boundaries, authn/authz, injection, secret exposure, filesystem and network reach, dependency risk. Use when the user asks for a security review, audit, or vulnerability check of concrete code. Not for general code review, lint, or compliance paperwork.
- ▌ Skill Installer · hmbownInstall, update, trust, or inspect Codewhale skills from GitHub or local skill folders. Use when the user asks for available skills or wants a community skill installed.
- ▌ Dependency Update · hmbownRead release notes/changelogs, update a defined dependency scope, handle breaking changes, and verify. Explicit-only; no broad update-everything by inference.
- ▌ Whalewiki · hmbownBuild and maintain a repository wiki with source citations, page and source hashes, conservative freshness checks, and an offline reader.
- ▌ Recording · hmbownCapture screenshots on registered computers, record on macOS or HarmonyOS, and manage saved captures. Probe capabilities before recording.
- ▌ Computer Use · hmbownDesktop control with accessibility-first observation and actions, pixel fallback, screenshots, zoom, screen recording, and switching between registered computers. Qualified on macOS; Windows, Linux and HarmonyOS backends are experimental.
- ▌ Cloudflare Docs · hmbownLook up current Cloudflare documentation (Workers, Pages, KV, R2, D1, Queues, DNS, WAF, platform APIs) through Cloudflare's official remote MCP server. Use when answering Cloudflare questions or writing Cloudflare config.
- ▌ Lava Stage · hmbownUse Lava Stage when the best move is to convert workable terrain into punishing terrain, forcing the opponent to either keep taking damage, leave valuable panels, or route through narrower safer lanes.
- ▌ Long Sword · hmbownUse Long Sword when the problem is visible locally but the fix must extend further — across file boundaries, into deeper call stacks, or through intermediary layers that a normal Sword cannot reach.
- ▌ North Wind · hmbownUse North Wind when progress is blocked by protective layering rather than distance: aura, barrier, wrapper, soft shield, ceremonial safety language, or stacked cover that keeps the real issue from being touched. North Wind strips the protection cleanly so a truthful next move can land.
- ▌ Panel Grab · hmbownUse Panel Grab for narrower room control when a single lane, dependency, or decision surface matters more than broad expansion.
- ▌ Sand Stage · hmbownUse Sand Stage when the best move is to change the board into slow, absorbent terrain that eats momentum instead of carrying it. Sand Stage is for bogging the field down: reducing clean repositioning, making extraction costly, and forcing both sides to respect how hard it is to cross or abandon the wrong patch once the ground starts swallowing motion.
- ▌ Slow Gauge · hmbownUse Slow Gauge when the room is moving too quickly to reason safely and the right move is deliberate rate control.
- ▌ Step Sword · hmbownUse Step Sword when the target is identifiable but not within easy reach: you know exactly what needs cutting, but reaching it requires navigating across files, modules, or abstraction layers. Warp directly to the target and make the edit in a single committed action.
- ▌ Wide Sword · hmbownUse Wide Sword when the problem is not a single precise target but a line of related issues sitting side by side. Where Sword cuts one thing cleanly, Wide Sword addresses a horizontal band of similar problems in one motion.
- ▌ Anti Damage · hmbownUse Anti Damage when the cleanest move is to convert incoming pressure into a reactive counter instead of tanking it.
- ▌ Full Custom · hmbownUse Full Custom when the current option set is depleted, stale, or wrong for the room and the best move is a clean redraw.
- ▌ Grass Stage · hmbownUse Grass Stage when the best move is to make the operating surface broadly hospitable to regrowth, healing, and low-friction expansion, while staying honest that the same lush terrain can amplify the wrong kind of pressure.
- ▌ Iron Shield · hmbownUse Iron Shield when a public-facing surface, endpoint, or interface needs durable, ongoing protection against repeated attacks on the same vector. Unlike a one-time absorb or a brittle guard, Iron Shield maintains a sustained defensive posture that endures multiple impacts from the front.
- ▌ Leaf Shield · hmbownUse Leaf Shield when you need sustained, active protection that makes aggression costly. The defense orbits your work and absorbs incoming damage; when triggered, it launches outward as a measured counter.
- ▌ Poison Face · hmbownUse Poison Face when Hermes should emplace a stationary poison source on the field: a bounded, persistent hazard that immediately taxes presence in one valuable lane, panel cluster, or holding pattern. Unlike Time Bomb, it is not waiting for a delayed detonation; unlike Killer Eye, it is not watching for a crossing; unlike Anti Damage, it does not need to be hit first. Poison Face is for making a known position steadily worse until the target leaves, reroutes, or accepts attrition.
- ▌ Poison Mask · hmbownUse Poison Mask when the right move is not to poison one tile on the floor but to contaminate a specific interface, workflow surface, or operating persona that the target keeps wearing. Poison Mask attaches the cost to continued contact: every repeated use of the same channel, wrapper, queue, ritual, or decision surface becomes more unpleasant, slower, riskier, or more draining. Unlike Poison Face, which punishes staying in one place, Poison Mask punishes continuing to operate through one poisoned surface. Unlike Time Bomb, it is not waiting to go off later; unlike Killer Eye, it is not there to detect passage; unlike Anti Damage, it is proactive rather than retaliatory.
- ▌ Under Shirt · hmbownUse Under Shirt when a catastrophic failure is possible and the system needs a last-resort failsafe that prevents total collapse, leaving it barely functional instead of dead.
- ▌ Curse Shield · hmbownUse Curse Shield when Hermes should protect a surface in a way that makes contact linger on the attacker. The shield is not just there to absorb the next hit or throw an immediate counter; it tags hostile touch with a lasting penalty, mark, or contamination that follows the offender after impact. In practice, Curse Shield is for defended surfaces that impose audit burden, throttling, suspicion, reduced trust, quarantine, follow-up scrutiny, or another durable consequence on whatever keeps pressing into them.
- ▌ Black Barrier · hmbownUse Black Barrier when ordinary buffering is too soft and thresholding alone is too permissive. The problem is not just one incoming hit, or a stream of weak noise, but contact that should have to prove itself before it reaches the work at all: ambiguous requests, half-authorized changes, untrusted touchpoints, speculative interruptions, or access that has not earned the right to cross the perimeter. Black Barrier establishes a denser selective shell around a critical surface. Unlike Barrier, it is not a one-hit absorb. Unlike Aura, it does not merely ignore weak impacts below a threshold. It enforces a stricter admission rule: only declared, validated, high-confidence contact gets through, and everything else is refused, deferred, or held outside until it can justify entry.
- ▌ Fighter Sword · hmbownUse Fighter Sword when the change must run across an entire line — a full sequential chain, pipeline stage, import row, or dependency thread — and anything short of cleaving the whole row leaves the job incomplete.
- ▌ Variable Sword · hmbownUse Variable Sword when you know something needs to be cut but the attack pattern — narrow precision, wide sweep, or long reach — should be determined by the shape of the target at execution time.
- ▌ Fan · hmbownUse Fan when the right move is short-range airflow control: gather loose material, pull a target toward a chosen lane, or create a directional repositioning effect without claiming territory or committing to a heavier strike.
- ▌ Aura · hmbownUse Aura when the problem is not one big blow but constant low-grade interference: nuisance traffic, shallow criticism, repetitive edge pokes, cheap probes, or background turbulence that keeps stealing turns. Aura sets a named threshold and blanks everything below it, so Hermes can keep operating through minor contact while preserving attention for impacts that actually matter. Unlike Barrier, Aura is not a one-hit absorb; unlike Leaf Shield, it does not punish contact. It is the baseline selective shell.