github.com/0gsd/enough
| Skill | Added | Review |
|---|---|---|
analyzer defaults/skills/analyzer/SKILL.md Four-mode analytical toolkit for documents, books, web pages, decisions, and untrusted skills. SUMMARIZE produces an even-handed one-page digest of any text (author motivation, intended audience, tone, key quotes). PROOFREAD makes light spelling/typo corrections on full-length documents (including entire books, chapter by chapter), driven by the Harper grammar checker (local, offline, rule-named findings), and produces a separate proof report with suggestions and repeated-phrase findings. DECIDE picks three archetypal personas from a built-in roster and runs a transcripted debate over a question, complex decision, or 'what should I do next' — returning a recommendation plus the full debate. AUDIT vets a third-party skill, readvisor, or paradigm before the user trusts it: a deterministic payload scan of the bundled scripts plus an LLM audit protocol against a threat model (prompt injection, privilege escalation, epistemic contamination, executable payload, audit evasion), producing a plain-English explainer, a pass/flag/fail verdict, and a report. Use for 'analyze this', 'summarize this', 'one-pager', 'tl;dr but smart', 'read this for me', 'proofread this', 'copy-edit this book', 'fix typos in', 'spot repetition', 'help me decide', 'pros and cons', 'what should I do about', 'three perspectives on', 'debate this', 'scan this skill', 'is this skill safe', 'audit this', 'vet this', 'check this before I enable it', 'what does this skill actually do'. Works with plain text, markdown, html, pasted text, web pages, free-form questions, skill folders or .skill zips, and any document enough converts to a markdown twin — Word documents, PDFs, EPUB ebooks, PowerPoint decks, Excel workbooks (read the twin, or read_file the original and enough converts it; PDFs, decks and workbooks need the PDF extra). | — | |
anything-finder defaults/skills/anything-finder/SKILL.md Deep-search retrieval engine plus patent prior-art search plus honest business-viability reads. FIND locates hard-to-find things a plain web search won't surface: public-domain texts; rare / lost / out-of-print film, TV and video (legal watch links); cleared / PD / CC0 images for covers, posters, zines; hard-to-find products, gear, synths and instruments (buy links, or an honest closest-thing / market-gap answer); paywalled or buried articles and papers (legit open copies — preprints, repositories, archives); open-source repos by permissive license (MIT / Apache-2.0), including libraries off GitHub; novel read-alikes; plus sheet music, samples, fonts, 3D models, stock footage, datasets, public APIs, hardware manuals, PD comics. PATENTS runs a structured prior-art search across granted patents, published applications, and non-patent literature, and reports novelty and non-obviousness concerns with an explicit not-legal-advice disclaimer. VENTURE composes the market sweep, the prior-art check, and a competitive-landscape sweep into an even-handed viability read — what exists, what's adjacent, what's genuinely open, the strongest case for and against building it. Trigger on 'find me...', 'where can I get / download / watch...', 'does there exist a...', 'track down a...', 'I'm looking for...', 'is there an open-source / free / public-domain...', 'prior art', 'patent search', 'has this been patented', 'is my idea patentable', 'novelty search', 'freedom to operate', 'is this a business', 'should I build this', 'does this exist as a product', 'evaluate this idea', 'market gap', or any request to locate a rare, buried, license-sensitive, or format-specific thing — or to find out whether an idea already exists in the world. | — | |
girraph-merirmaid defaults/skills/girraph-merirmaid/SKILL.md Two sibling visualization jobs in one skill. GIRRAPH — IBIS discipline for mapping contested, fuzzy, or wicked problems as issues (❓), positions (💡), and arguments (➕/➖) in a .girraph file, BEFORE any solving happens; engage when the user wants to map a problem, untangle a decision with no clean answer, see the shape of a disagreement, or says 'girraph this', 'map this out', 'help me think through (not solve) X', 'what are the positions on Y', 'IBIS', 'argument map'. MERIRMAID — generate, display, and edit Mermaid diagrams in .merirmaid files for ANYTHING with structure or flow; engage when the user says 'diagram this', 'flowchart', 'mermaid', 'sequence diagram', 'state diagram', 'visualize this process', 'draw the architecture', 'show how X works', 'chart the steps'. Also engage when a planning conversation keeps circling — competing considerations that won't linearize are a girraph smell; a described-but-unpictured process or structure is a merirmaid smell. Do NOT engage for problems with a known procedure (use a request file), linear plans (use text-planning), or simple either/or questions the user just wants answered. | — | |
lexicographer defaults/skills/lexicographer/SKILL.md Looks words up in FEED, the first-party enough english dictionary, and writes the user's own dictionary entries with them — a coinage, a family word, a term from their field, a private name for a thing, or their own version of a word FEED already has — one column at a time, in FEED's house style. Use for "is X a word", "what does X mean", "look up X", "add X to my dictionary", "put this word in the dictionary", "I made up a word", "my family says X", "fix the entry for X", "my version of X", "define X for me and keep it". Not for translating text (that is the translator skill) and not for spelling or grammar checks of a document (that is analyzer's proofread). | 76 Impact — No eval scenarios have been run Securityby Passed No findings from the security scan Version: 3cf8a60 | |
macrocosm-oes defaults/skills/macrocosm-oes/SKILL.md Reads, checks and explains MacroCOSM `.cosm` files — Holotypes, Canon catalogs, product trees, fabulas, holographs — and, only when the user asks, decompiles a text into its fabula, writes its atem (the tightest fair, full, plain reduction of a text), puts a volume's questions to the user and records their answers, builds a fabulac of a folder of Python, or runs a Thym program. Use for "what is this .cosm file", "read this Holotype", "lint this .cosm", "decompile this text", "make an atem of this", "say it in one line", "what is a fabula", "what does this stance mean", "ask me the questions", "build a fabulac", "run publish-story on this", "what is Thym". Not for an ordinary summary in the user's own terms (answer that directly), not for editing a Holotype's content (that is Comprise's work), and not for writing new .cosm volumes by hand. | 74 Impact — No eval scenarios have been run Securityby Passed No findings from the security scan Version: 3cf8a60 | |
memoir-dialectic defaults/skills/memoir-dialectic/SKILL.md Multi-session memoir planning and (optionally) drafting via patient, iterative dialogue. Use whenever the user wants to plan, build, or write a memoir — full life, partial life, single milestone, professional thread, creative thread, or thematic slice. You interview the user across many sessions, accumulate plan documents in a project folder, and optionally synthesize a draft. Trigger on memoir-dialectic, "help me plan my memoir," "write my memoir with me," "memoir interview," "life story project," "autobiography help," or any long-horizon autobiographical writing collaboration. | 74 Impact — No eval scenarios have been run Securityby Passed No findings from the security scan Version: 3cf8a60 | |
readvisory defaults/skills/readvisory/SKILL.md Builds a new readvisor out of a real person's judgment, by interview, and installs it. Two ways to gather: LIVE — a patient interview of the user themselves, one or two questions at a time, twelve to eighteen in all; or QUESTIONNAIRE — a plain, email-ready file the user sends to someone whose advice they want on hand (a friend, a mentor, a colleague, a parent, a former editor), with the answers pasted back whenever they arrive. Always ends the same way: a follow-up pass to the user, a preview of both documents they can correct line by line, and their go-ahead before enough installs anything. Produces AGENT.md and MOTIVATION.md and installs them with the install_readvisor tool, in this project or on this machine. Use for 'make me a readvisor', 'build a readvisor', 'new readvisor', 'readvisory', 'interview me and turn it into a readvisor', 'I want my friend's judgment in here', 'profile how my old editor thinks', 'send my dad a questionnaire', 'turn these answers into a readvisor', 'I want a second opinion that argues like X', 'someone who will push back on me about Y', 'I miss asking her about this'. Not for renaming the chief readvisor, not for running a council — this makes the participants, it does not convene them. | — | |
scaffold defaults/skills/scaffold/SKILL.md Turns a brain dump into a visible structure. Reads a pile of thinking — a story idea, an essay, an argument, a plan, a proposal, a talk, a course — and returns a markdown outline that enough converts deterministically into a composure the user can look at, move around, and fill in. Knows story shapes (arcs, beats, continuity threads, denouements, endings) and non-fiction shapes (claim, grounds, warrant, counter-case, close; goal, phases, dependencies, risks, done-when). Never invents material to fill a hole — it writes a [gap: …] card instead, so the structure shows the writer what they still owe it. Asks at most two clarifying questions before the first pass. Works from text pasted in chat, a file in the project, or an existing composure. Use for 'scaffold this', 'give this a structure', 'structure my notes', 'outline this', 'turn my notes into a composure', 'make me a storyboard', 'map out my essay', 'what shape is this', 'break this plan into phases', 'lay out the acts', 'I have a pile of notes and no structure', 'where are the holes in this', 'I know what happens but not in what order'. | — | |
translator defaults/skills/translator/SKILL.md Offline translation across ~419 languages via MADLAD-400-3B-MT (CTranslate2 + SentencePiece). No cloud. No account. The model downloads once (~3 GB) and runs locally on CPU or Apple Silicon thereafter. Use whenever the user wants to translate text TO or FROM a non-English language — short phrases, paragraphs, whole documents, idioms, low-resource languages, indigenous languages, dead languages MADLAD has seen, polite forms, casual registers. Also trigger on "translate", "say this in <lang>", "what does <foreign text> mean", "render this in <language>", "<lang> for <phrase>", "how do you say X in Y", "localize", or any request that involves moving text between human languages. DO NOT use for code translation between programming languages. DO NOT use for transliteration (script conversion) without a target meaning — MADLAD translates meaning, not glyphs. | 65 Impact — No eval scenarios have been run Securityby Passed No findings from the security scan Version: 3cf8a60 |