Generate editorial-style HTML information cards and capture them as ratio-specific PNG screenshots. Use when the user wants a research card, info card, briefing card, cover card, or editorial social asset in ratios such as 3:4, 4:3, 1:1, 16:9, 9:16, 2.35:1, 3:1, or 5:2. Includes stable Chinese typography fallback and…
A multi-review writing workflow that checks text through six separate lenses, including facts, unnecessary length, structure, word choice, tone, and rhythm. It then combines the findings, rewrites the text, and checks the result again.
A read-only review agent that scans a source document through one assigned review lens and records detected problems in a digest file. It diagnoses issues but does not rewrite the document.
An optional fact-checking agent for precise claims such as numbers, dates, prices, versions, policies, results, and market information. It verifies those claims with external sources and reports what it finds without rewriting the text.
An independent review agent that compares an original text with its rewrite. It checks whether facts, conditions, examples, meaning, or necessary explanations were lost or changed.
A coordinator for a writing-quality review process. It assigns separate reviewers to inspect different problems, sends rewriting to another agent, and uses a verifier to check the result.
An editing assistant that applies a prepared list of writing problems to an existing document. It makes the smallest changes needed while preserving facts, numbers, names, dates, quotes, and other fixed details.
An optional agent that studies samples of an author's writing and produces a style guide for another agent. It is used only when the user asks for voice matching and supplies two to five representative passages.
A skill for analyzing technical community chat records from text, screenshots, JSON, or CSV. It creates member profiles, activity assessments, influence rankings, and community-health findings.
A skill for choosing and creating clear charts from user-provided data, following the Storytelling with Data approach. It matches chart types to tasks such as comparing categories, showing trends, or displaying distributions.
A writing workflow for producing structured, journal-style laboratory, research, engineering, or competition reports from experiment details and results.
A framework for coordinating several AI agents on a complex task by planning first, assigning isolated work areas, and combining their results. An agent is a separate AI worker given a specific part of the job.
Initialize or upgrade a project to the full README First architecture from endearqb/ReadmeFirst. Use only when the user asks to initialize README First, install the README First framework, create the complete AGENTS.md + README.md + .ai/changes + .ai/decisions structure, or retrofit a repository with the README First…