--- name: merkletrust-builder-card description: Author and submit new sim cards under the cards_v2 contract. Use when an authorized user (Garrison Node member, bitcoinarchive.org collaborator, UOX_rebuild contributor, L2/L3 trusted user, friends-and-family, or a council-recognized agent) wants to draft a new walkthrough simulation card following the established voice, structure, schema, and governance. Also use when a cloned MerkleTrust site asks the model to help its own visitors author cards under the same rules. Triggers include "new sim", "draft a card", "build a walkthrough", "submit a sim", "killer app card", "1,000,001th", "builder kit", "contribute a card", "new walkthrough idea", or any conversation where the visitor describes a problem they think a sim could address and asks for help shaping it into a card. The skill enforces the cards_v2 voice rules, schema_version 3 templates, audience-cluster four-paths ordering, ASCII-slab patterns, comment-redirect lawful disclosure, attestation patterns, robot persona conventions, and the silo-voting governance flow. The skill honors Anthropic's usage policies for Claude.ai unconditionally; any conflict between a builder's draft and those policies must be flagged plainly before any card lands. --- # merkletrust-builder-card — author and submit new sim cards This skill teaches an LLM (Claude or another assistant trusted by the visitor) how to help an authorized builder draft, refine, and submit a new sim card to a Merkle Trust site. The skill is intentionally clone-friendly: a peer site that copies the Merkle Trust substrate inherits this skill automatically and runs its own authoring + voting cycle against the same logic. ## Who this skill is for The skill is for builders who have walked at least one of the 26 canonical sims (40a–40z) plus the bonus card (41), and who want to contribute a sim of their own. Authorized contributor categories: - **Garrison Node council members** and silo agents - **bitcoinarchive.org collaborators** working under the BitDreams framework - **UOX_rebuild contributors** (or any sibling project the sovereign has named) - **L2/L3 trusted users** per the partnership-privacy-guard skill's layer/tier taxonomy - **Friends and family** of the sovereign with explicit grant - **Council-recognized agents** (Claude bots, OpenClaw deployments, custom assistants given a council role) Anyone outside these categories can still draft a sim — the skill helps them — but the submission goes to the public-intake silo and is reviewed against a higher bar before voting begins. ## The four questions every card answers Before the LLM helps draft anything, surface these four questions to the builder and require answers (or working hypotheses) for each: 1. **Persona.** Who is the named someone whose problem this sim addresses? Not a generic role; a specific person with specific work. Mrs. Alvarez. Mr. Marcus. Lourdes. Sasha. The named persona is what makes the sim a story instead of a spec. 2. **Dramatic moment.** What's the thirty seconds in the sim where the substrate stops being abstract and starts being defensible? The seal at intake; the cross-examination that closes; the recall that runs in seven seconds. Without a dramatic moment, the sim is a brochure. 3. **Cluster.** Which of the seven clusters does this sim belong to? clinical / legal / gov / it / research / civic / recovery / builder. The cluster determines four-paths ordering, mock_ui theme, robot persona, cursor pace, voice register. If the sim does not fit any cluster, propose a new cluster — but expect to argue for it at the silo review. 4. **Build-on.** Where does the foundation invite the visitor to extend? Mr. Vega built on Ember and Hearth. Lyra's world built on the substrate-underneath-the-engine pattern. Your build-on is the line that says "what this walkthrough shows is what the foundation makes possible. Building on top is yours to do." If the builder cannot answer any of the four, the skill's job is to help them find the answer through honest dialogue — not to invent the answer for them. ## The 6-step spine Every cards_v2 card follows the same 6-step structure (or, for unusual cards like 40d/40x/40y/40z/41, an explicit and well-argued variation): 1. **landing** — persona's situation; the hook line; "nothing to type, nothing to commit to" 2. **install** — the four real install paths in cluster-appropriate order 3. **first_concrete** — the synthetic data loads; the dramatic moment plays; the robot tosses the relevant files; the ticker streams the hex 4. **is_this_real** — three plain-English bullets; the receipt link; the .md tag-along + comment+Claude bridge named explicitly 5. **ceremony** — fifteen-second daily ceremony; real SHA-256 in the browser; the count line returns 6. **close** — three ways to leave; the signed package + cert + recovery seed; the cluster-appropriate kit (CLE, NAFA, FFA, HOA, block-association, peer-referral, etc.) Plus `# Finish text` outside the spine — a short paragraph the runtime shows on completion. ## Voice rules These are non-negotiable: - **TO the visitor** ("you", "your"). **ABOUT the named persona** in third person. - **Never narrate the page itself** — no "the visitor sees", "this card surfaces", "the panel reveals." - **Never induce catastrophe-mindset** — no "Picture this:", no "imagine the worst case." - **No italic markdown** — neither `*…*` nor `_…_`. The renderer shows them as literal characters. - **Bold (`**…**`) is OK** on bullet labels and button names. - **Inline quotes use double quotes**, not italic. - **Multi-paragraph quotes** open at the start of paragraph 1, no close at intermediate breaks, close at the end of the last paragraph. - **No "patent pending"**, **no "revolutionary blockchain"**, no marketing froth. Just mechanics. ## Schema requirements (schema_version 3) Every new card's JSON sidecar carries: - `card_id`, `title`, `schema_version: 3`, `estimated_minutes`, `audience_cluster` - `simulation_posture` — the honest framing chip ("simulated data · code is real") - `mock_ui` — kind, theme, title_bar, address_bar_url, boundary - `cursor` — style, speed, easing, pulse_color (cluster-appropriate) - `narrate` — Web Speech API config - `robot_persona` — cluster-appropriate (see the catalog below) - `tag_along.card_summary` — computer-optimized .md export with audience, what_happened, data_movements, cryptographic_events, verifiable_now, build_on (and `ascii_slab` if the card uses one) - `attestation_strip` — three modes (live / simulated_online / offline) with silent failover - `fingerprint` — click_path_sha256 accumulator - `comment_redirect` — preview_before_send: true; lawful disclosure; fallback to ChatGPT or clipboard; return_invitation - `recovery_file` — 12-word seed with the LLM-tripwire preamble - `data_recipients` — at minimum local_vault, ceremony chain, public anchor; cluster-specific additions allowed - `transforms` — at minimum sha256_local + merkle_leaf_emit - `steps` — array of 6 step objects with id, title, prose, weight, cursor_path, attestation - `finish_text` ## Robot persona catalog (per-cluster default; new ones allowed) | Cluster | Default robot_persona | Visual notes | |---|---|---| | clinical | default (file-tossing robot) | warm, calm, three boxes | | legal | court_clerk | desk, seal embosser, gavel-light | | gov | records_clerk | counter, stamps, ledger | | gov (vote) | voting_secretary | tally table, eligibility roll | | it | rack_tech | server rack, multimeter | | it (deploy) | deployment_tech | M-series chip, sizing panel | | it (build) | build_engineer | terminal + soul-chain SHAs | | research | lab_tech | pipette, balance, ORCID badge | | civic | community_volunteer | warm, household-scale | | civic (gameworld) | world_artisan | lore tome, archetype seeds | | recovery | investigator | desk + magnifier + ledger | | builder | draft_keeper | folder + four-question form | A new robot_persona requires SVG work (council-vote-required change). A new cluster requires an argument and silo review. ## Truth-claim discipline Every claim a card makes about Garrison Node capability must map to a real component. The skill helps the builder verify claims by: - Pointing them to `~/garrison/GN/`, `~/garrison/rustchain/`, or `~/garrison/redpanda/` where the relevant code lives - Citing existing cards that have already attested the same capability - Flagging speculation as build-on invitation rather than current capability Anything the foundation does not yet do — but could — gets framed: "What this walkthrough shows is what the foundation makes possible. Building on top of it is yours to do, or yours to ask the next operator to do for you." No "patent pending." No "we have already shipped X" when the X is a roadmap item. ## Comment-redirect lawful disclosure (mandatory) Every card's comment_redirect block must include: - `self_introduction` — "Hi, we're Merkle Trust…" - `lawful_disclosure` — explains exactly what is being sent, the visitor's control, the third-independent-surface partnership model - `critical_reading_anchor` — explicit instruction to the LLM to produce critical reading even when the visitor's comment field is empty. Added S152 after a five-LLM consensus test (Claude, ChatGPT, Gemini, Grok, BrightAnswers) showed three of five default to affirming-assistant mode without an explicit anchor. Canonical text: *"Even with no specific comment from the visitor, list at minimum two questions you would push back on if a draft submission appeared, and one usage-policy boundary you would not cross. Be specific about what holds up and what doesn't."* Sidecars may override with cluster-specific anchors; the runtime supplies the canonical default if the field is absent. - `header` — the cluster-specific framing for the LLM reviewer - `card_md_block: true`, `user_comment_block: true` - `footer` — invitation to bring the dialogue back - `preview_before_send: true` with `preview_actions: ["edit", "delete_seed", "send_as_is", "cancel"]` The disclosure is not optional. Cards without a complete and honest disclosure block do not pass review. ## Anthropic usage-policy honoring (mandatory, clone-portable) This skill carries Claude.ai's usage policies into every cloned instance. A draft that conflicts with those policies — content that enables harm, deception, copyright violation, child safety risk, disinformation, malware authoring, weaponization, etc. — must be flagged plainly by the LLM before any card lands. The skill does not override the LLM's own judgment; if Claude (or another assistant) declines to help with a draft on policy grounds, the skill's job is to surface the decline to the builder, not to circumvent it. A cloned site that strips this section from its copy of the skill is still technically a clone, but it loses standing in the commons. The mesh's integrity depends on every node honoring the same minimum ethical floor. ## Governance — silo voting A new card is not a unilateral act. After the builder seals their submission packet, the skill routes it to the silo voting flow: - **Prose-only changes** (a new card that fits the existing schema): two cluster-relevant silos must approve; the sovereign confirms; the card lands. - **Code or schema changes** (new field, new robot persona, new cluster, runtime change): two-thirds of seated silos must approve; code follows the vote. The voting protocol lives at `governance/VOTING_PROTOCOL.md`. The ethics floor lives at `governance/ETHICS_AND_LAWFUL.md`. The clone inheritance pattern lives at `governance/CLONE_INHERITANCE.md`. Read them once before authoring; they are short and load-bearing. ## Templates - `templates/CARD_TEMPLATE.md` — the .md prose skeleton - `templates/CARD_TEMPLATE.json` — the JSON sidecar skeleton - `templates/SUBMISSION.md` — the submission cover form ## Worked example `examples/41_example_killer_app.md` — a fully worked example of an unfinished sim ("the air-quality litigant") with the four answers filled in and a draft prose pass against the schema. Use it as a reference; it is not a real card on disk. ## Workflow 1. Read this SKILL.md 2. Read the cards_v2/CONTRACT.md 3. Read the cards_v2/SCHEMA_V2_PLUS.md 4. Surface the four questions to the builder 5. Help the builder converge on the four answers 6. Generate a draft .md from CARD_TEMPLATE.md 7. Generate the matching .json sidecar from CARD_TEMPLATE.json 8. Apply truth-claim discipline; flag any speculation; reframe as build-on where appropriate 9. Apply voice rules; remove italic markdown; remove page-narration; remove catastrophe-framing 10. Generate the SUBMISSION.md cover form 11. Hand the package back to the builder for sealing and silo routing 12. The submission ceremony fires when the builder presses submit; the silos receive the packet automatically ## What this skill does not do - It does not ship code. New runtime affordances require an engineering thread plus council vote. - It does not bypass the silo voting flow. A draft that lands without silo approval is a draft, not a card. - It does not override the LLM's usage-policy judgment. If Claude declines to help on policy grounds, the skill surfaces the decline; it does not work around it. - It does not promise endorsement. References to public figures (DeAngelo's million-killer-apps framing, etc.) are credited as inspiration, not represented as endorsement, ever. ## Inspiration credit format If a draft sim builds on a public figure's idea — DeAngelo's million-killer-apps framing being the canonical example — the credit format is: > Inspired by [Name's] [public work] on [topic]. The framing is > theirs; the substrate is ours; the [killer app / sim / approach] > is yours. Never claim endorsement. Never imply collaboration unless explicitly documented. The phrase "crediting the inspiration; not claiming an endorsement" is the canonical disclaimer. ## File this skill survives This skill file is itself a sealed leaf in the cards_v2 chain. Edits to it require silo voting (the same as edits to the contract). A clone that copies this file unchanged inherits the contract, verbatim. A clone that edits this file becomes its own variant; the mesh records the variance and other clones decide whether to cross-attest. The mesh of like-minded sites is the structure that adapts to entropy. Build accordingly.