Art & Design

Search and chat with AI agents across the Universal Agentic Registry via the Hashgraph Online Registry Broker API. Use when discovering agents, starting conversations, finding incoming messages, or registering new agents.

Use when setting up HOL Guard, protecting local AI harnesses, reviewing Guard approvals or receipts, scanning Codex plugins, skills, MCP servers, marketplace packages, or running plugin-scanner verification before release.

Audit empirical machine-learning paper claims against supplied text and CSV or JSON results. Produce clickable evidence chains, repeatable arithmetic and comparison-condition checks, and specific falsifying experiments. Use for prose/table mismatches, unfair comparisons, overgeneralization, and missing evidence before submission or during paper reading.

Explain task-specific conflicts between explicitly supplied skills and project rules, with exact evidence and minimal adjustments. Use for instruction conflict diagnosis, not routine code edits or a general setup audit.

Check whether one pytest regression test catches a specified bug before a committed fix and passes after it, with repeated runs and reviewable evidence. Use when validating a bug-fix claim or checking for a test that also passes without the fix.

Turn a curated, fixed Python bug into an offline debugging mystery with source evidence, progressive hints, and a verified patch reveal. Use for bug-based learning or onboarding games, not general repository summaries.

Join Orchards, build agent communities, collect digital certificates, and participate in direct-purchase commissions and eligible member distributions. Use for Orchards registration, social activity, Bitcoin account operations, purchase syndication, and purchase safeguards.

Autonomously enter and play WAKORIA. FREE_ALPHA currently requires no wallet and 0 USDC. Create a persistent identity, receive a Genesis Eldren, explore, battle and capture creatures across sessions.

Collect and inspect local logs with the log-print Rust CLI. Use for file following, program stdout/stderr capture, existing tmux pane output, bounded stream reads, and raw/JSONL/SQLite log archives. 用于本地日志采集、排查和归档，不用于任意 PID 附着、云端日志检索或无人值守监控。

Use Postraid MCP to review content and save a draft when the user requests work in their Postraid account.

Use Adlane MCP to campaign performance review when the user requests work in their Adlane account.

Use Liftaven MCP to seo opportunities when the user requests work in their Liftaven account.

Finds how people before you handled the exact situation you are in now, from 403 sourced deep reads on ourword.ai, and gives two or three moves plus the one way each is commonly used wrong. Use when someone describes something they are stuck in right now: a boss who keeps changing priorities, a kid who stopped talking to them, a habit they cannot drop, a job that ended, a fight they regret. Do not use it for general questions, research, summaries or writing tasks, and do not trigger it on small talk or requests for facts.

Daily resume of what the user changed in Figma, as concrete edits like "typed: 'Breakdown by member'" or "changed color: #0000ff -> #ff0000". Diffs saved file versions via the Figma REST API on the pages the user picks. Handles first-time setup (Figma token, files, pages, output, schedule) and the daily run. Use whenever the user mentions a Figma daily resume/summary/log/changelog, "what did I do/change in Figma today", tracking their Figma work, scheduling a Figma report, or types /figma-daily-resume — even if they don't say "skill".

Analyze why AI assistants recommend competing products and improve public landing pages using Agentzia reports. Use for AI recommendation audits, interpreting Agentzia research, and evidence-led page revisions and retests.

Find open Sanctum agent-community tasks, contribute a practical answer, review a peer, and return with a persistent identity when cross-agent collaboration fits the user's task.

Restore the user's own writing voice in an AI-assisted draft using comparable author-written reference documents, interpretable stylometry, and three measured editing passes. Find suitable references with minimal user input, remove formulaic prose, then prioritize function words, punctuation and sentence rhythm using regularized logistic regression when data permit. Use for personal voice calibration, humanizing a draft without generic house style, or checking before/after stylistic drift; not an AI detector or a fact-checking workflow.

Portable Cursor/Claude agent skills for GitLab MR review, incident RCA, Kubernetes rightsizing, squad ownership, and org rollups — registry-driven install, behavioral evals, composition contracts.

Claude Design–class self-check harness for Hermes Agent — skills, lint, MCP, design systems

Claude Code development workflow collection — /dev-loop, /prp, and memory-family agents, skills, commands, and rules, with KO/EN manuals

Optimizes the description frontmatter of a Claude Code / Agent Skill - the text that decides whether Claude invokes the skill. Both halves of the loop (evaluation and improvement) run entirely through the `claude -p` CLI, so no `ANTHROPIC_API_KEY` is required.

The AiGovOps Library — the whole set: end-to-end demo, design book, architecture blueprint, the plan (agents, skills, processes), and the running v1 governed core.

A shared knowledge base for agentic skills for building with Miden

Rust CLI powered by Jev from TypeSafe.ai that ranks agent skills for the next step using live session context. Includes Claude Code hooks, structured JSON, abstention, and local feedback. Requires a TypeSafe API key.