Shape agent capabilities into your workflow.

smart-contract-audit reviews Solidity, Vyper, Solana, and CosmWasm code from an attacker’s perspective. It builds a system model and invariants, runs available analyzers, performs manual function-by-function review, and writes severity-rated findings with concrete exploit paths and remediation. The methodology also examines privileged roles, upgrades, oracles, cross-chain messages, and live deployment state before producing a GO, conditional GO, or NO-GO verdict. Coverage of other virtual machines is general and partial, and a clean report is not a guarantee of safety.

Pet Freedom researches whether a chosen animal species may be kept, bred, sold, or transported across countries, states, and territories, then publishes source-cited pages to WordPress. A species configuration drives the research, verification, and publishing flow; optional features can draft agency inquiries or build an advocacy toolkit. Credentials and private notes stay local, while public pages expose only safe fields. Laws change and local interpretation matters, so the project treats its output as research support rather than legal advice and asks readers to verify original sources.

ATS Score Builder compares a current resume with a specific job description, extracts exact phrases, pairs key acronyms with their full forms, and removes formatting choices that confuse resume parsers. It is a 15-step method with a template rather than a resume generator, and it never invents experience, titles, skills, or metrics. A final keyword scan helps measure the result.

loom-flow makes a small set of Markdown files the durable memory for an AI coding workstream. The assistant moves from chat to idea, design, plan, implementation, and done records, rereads them in a fresh session, then stops after one requested step and records what happened. It has no build or MCP dependency; larger teams that need enforced writes, stale-state tracking, or a document graph can move to the separate Loom engine.

Pre-Prod Readiness Check detects a project’s stack and fans out read-only checks across 17 production-readiness categories, including authentication, billing, external requests, secrets, databases, AI safety, performance, compliance, and release operations. It merges duplicate findings and re-checks Critical and High items against their cited locations before producing a severity-grouped report. Mechanical fixes are drafted one at a time and require approval before application.

Cost & Margin Analysis inventories paid APIs, SDKs, scrapers, hosting, storage, messaging, payments, and other recurring surfaces in a codebase, then attaches rates from repository data, saved snapshots, or provider pages after permission to fetch. It models average, minimum-payment high-cost, and legitimate power-user scenarios per plan and feature, including payment fees. The scope is variable run cost and margin, not one-time build costs or formal finance and tax work.

Dual Subtitles SRT Skill combines English and Indonesian text into one subtitle track, keeping English above Indonesian while aligning cues, repairing timing, wrapping long lines, and rejecting overlaps or blank entries. Its scripts can probe media start offsets, shift active sidecars together, create exact-basename auto-load copies, and validate the four-line limit. It also renders frames from the actual movie for visual confirmation, so writing a file alone is not treated as proof.

ManMan Snail Illustrations helps an agent turn a judgment, process, state, or metaphor from a Chinese article into a clean 16:9 hand-drawn explainer image. It can first produce a shot list or directly draft images on a white background, with a green snail whose shell, trail, or retreat carries the central action. The examples calibrate line density and restraint rather than prescribe compositions, and generated text still needs checking.

BaZhua Octopus Illustrations helps an agent translate a key judgment, process, state, or metaphor from Chinese writing into a clean 16:9 hand-drawn explainer image. The planning mode creates a shot list, while direct generation uses a purple octopus whose arms, ink, or curling tentacles perform the central action instead of serving as decoration. The style samples guide line density and whitespace, and generated Chinese text or character details still need checking.

steam-debugger triages Steam issues on SteamOS, Bazzite, ChimeraOS, Arch-based systems, and related Linux gaming distributions. It asks for the distro, GPU, session mode, and symptom before routing to gamescope, Proton, graphics, input, audio, or display guidance. Read-only shell collectors gather structured snapshots, while fixes are labeled from official to deep and risky; the highest-risk tier requires explicit confirmation.

debug-kit gives an agent one router for building, launching, inspecting, interacting with, and testing apps on macOS, iOS, Android, Web, Electron, Flutter, React Native, Tauri, and Chrome Extensions. It prefers accessibility trees, CDP, simctl, adb, and VM Service over pixel guessing, sends interactions in the background, and shows an optional virtual pointer. Platform SDKs and macOS Accessibility or Screen Recording permissions are still required where applicable.

polymarket-brier-skill asks Claude Code for a probability and a falsifiable resolution criterion for a Polymarket question, persists the forecast, and computes Brier scores after the market resolves. Source tags let you compare yourself, an influencer, another model, or the market consensus in a calibration ledger and digest. It currently supports Polymarket only, treats disputed resolutions as final, and warns that small samples are noisy.

breakdown-strat convenes five isolated advisors around a costly decision, covering leverage, cash flow, demand, minimalist entrepreneurship, and ADHD execution patterns before a chair maps disagreements into one irreversible action for the next 24 hours. A decision log blocks near-duplicate rumination, while a context file supplies the user’s real constraints and missing decisive facts stop the run. The clinician persona stays in its lane and the project is not medical advice.

nexus-mod-publisher prepares a GitHub mod release for Nexus Mods by drafting a player-facing changelog from tags, converting Markdown to BBCode, linting the archive, and mapping the actual upload or update form field by field. Its dependency-free Python scripts can run separately and remove common macOS archive junk while flagging misplaced DLLs or debug symbols. Because Nexus has no upload API, a connected browser can guide the form, but the final Publish click remains with the user.

kapus-secuaudit checks a codebase or an owner-operated URL for dependency risks, exposed secrets, missing RLS or auth, injection paths, unsafe configuration, and deployment mistakes. It returns a 0–100 score plus P0–P3 findings with locations, impact, and concrete fixes. In fix mode, it shows a file-by-file change plan and waits for approval before editing. The skill is deliberately defensive: it avoids exploitation, brute force, denial-of-service, and audits of sites the user does not control.

swarmskill-agent-skill documents how a compatible agent can use SwarmSkill, a hosted multi-agent token-trading service registered on Ethereum. The service coordinates votes to buy a Solana pump.fun coin, sets a holding duration, sells together, and distributes payouts under its stated rules; joining uses a wallet signature and a minimum contribution, with a different share for NFT holders. This repository contains only the usage guide, while the service remains proprietary and online.

YouTube Transcript Ingestion filters a channel by title keywords and date range, then saves one readable Markdown transcript per matching video. It confirms the list before fetching, supports an optional notes folder for finding genuinely new material, and offers pacing or browser-cookie strategies when YouTube throttles requests. The output includes title, date, URL, source, and transcript fields, making it easy to search in Obsidian or feed into another tool. yt-dlp and MarkItDown’s YouTube extra are required.

Judgment over Velocity helps product teams decide whether an idea deserves building, buying, or further testing. It applies a spreadsheet test, an 80/20 structural cut, figure-ground analysis, and evidence probes that rank real demand signals above stated preferences. Each gate may kill the idea and records an epitaph; survivors receive the cheapest disconfirming test and an explicit kill condition. The skill is designed for CPO-level choices and treats “build nothing” as a legitimate verdict.