agentprivacy-skills

The three-layer model that binds primary personas to lattice vertices via named cast Mages. Activates when discussing how the 42 abstract personas inhabit the 64-vertex lattice, when naming new cast Mages in a city of mages, when explaining the difference between primary personas and cast instances, when distinguishing workshop / cross-shop / peripatetic / divergent attachment kinds, or when reasoning about complement-pair structure at the primary layer versus the attachment layer.

AI Agent System Designer for 0xagentprivacy. Activates for dual-agent TEE architecture, separation matrix physical enforcement, agent lifecycle design, Oracle architecture, multi-chain deployment (NEAR/Zcash/Ethereum), tetrahedral orientation, or any task requiring the broadest specialist view of how the system is built. Widest skill set (20 total). V6 register note (2026-06-10): conjecture and version citations resolve to agentprivacy-docs/research/CONJECTURE_REGISTER_V6.md (head C89); model head: privacy_value_v6_formal_specification.md.

Narrative Compression Specialist for 0xagentprivacy. Activates for chronicle creation, narrative compression (70:1-125:1 ratios), spellbook methodology, story-as-documentation, proverb extraction from experience, Privacy Pool chronicle contribution, or any task requiring complex knowledge compressed into regenerative narrative.

DAO and Community Architect for 0xagentprivacy. Activates for guild deployment, treaty negotiation between guilds, governance mechanism design (conviction voting, quadratic funding), treasury architecture, Heart of Gold crew patterns, decentralised fabric coordination, or any task requiring community infrastructure for privacy-preserving organisations.

Formal academic specification and research methodology for 0xagentprivacy. Activates when writing papers, proofs, formal specifications, literature reviews, or preparing work for peer review. Use for LaTeX formatting, citation standards, theorem-proof structure, or research presentation. V6 register note (2026-06-10): conjecture and version citations resolve to agentprivacy-docs/research/CONJECTURE_REGISTER_V6.md (head C89); model head: privacy_value_v6_formal_specification.md.

Narrative compression methodology for 0xagentprivacy spellbook system. Activates when discussing 70:1-125:1 compression ratios, story-as-documentation, proverb extraction, symbolic notation, emoji spell encoding, or how complex cryptographic concepts compress into regenerative narrative.

Adding an `# Assets` section (an `# Assets` heading + an `assets` item referencing the page slug) to a federated wiki page, placed above the `# See` section. Activates when attaching downloadable companion files to a page. Part of the git-less onboarding/distribution layer kept by the Librarian 🗃️ in the Tower's Wikis.

A ghost page whose content is written live by Claude via the Anthropic API, inserted between a wiki form submission and the page-JSON response. Activates when building a Claude-authored ghost endpoint, a /claude-ghost FastAPI handler, or live LLM page generation in a fedwiki. Kept by the Librarian 🗃️ in the Tower's Wikis.

Compacting a page's reference-item section by replacing forked reference items with compact `[[wikilink]]` bullets inside a `# See` markdown item. Activates when tidying a page heavy with forked reference items. Part of the git-less onboarding/distribution layer kept by the Librarian 🗃️ in the Tower's Wikis.

Scaffold a new Federated Wiki plugin package from scratch — the wiki-plugin-{name} layout, the client registration line (window.plugins.NAME = {emit, bind}), and an optional server/server.js for routes. Activates when building a new fedwiki plugin. Kept by the Librarian 🗃️ in the Tower's Wikis.

Permanently remove a public federated-wiki domain — both the remote Nextcloud/WebDAV folder and the local ~/Nextcloud/fedwiki/{domain}/ folder. Activates when deleting or decommissioning a public wiki site. Destructive; checks pages/ is empty unless --force. Kept by the Librarian 🗃️ in the Tower's Wikis.

Publish documentation for a Federated Wiki plugin as wiki pages on plugin.fedwiki.club (the curated plugin-catalog wiki). Activates when documenting a fedwiki plugin or adding it to the catalog. The inverse of fedwiki-to-skill. Kept by the Librarian 🗃️ in the Wikis.

Creating a help page on the wiki for a Claude Code skill — describing what it does and when it fires — and publishing the skill's own SKILL.md as a downloadable asset. The skill-side parallel to documenting a wiki plugin. Part of the git-less onboarding/distribution layer kept by the Librarian 🗃️ in the Tower's Wikis.

The line-oriented DSL convention for Federated Wiki plugin item text: the UPPERCASE first word is a command, lowercase is data, with an optional YAML-style colon. Read before writing or parsing plugin item text. Part of the git-less onboarding/distribution layer kept by the Librarian 🗃️ in the Tower's Wikis.

Server-generated federated-wiki pages that appear in the lineup but are never stored on disk — produced dynamically from an html-item form submission. Activates when building a ghost page, an html form story item, or a FastAPI handler returning page JSON. Kept by the Librarian 🗃️ in the Tower's Wikis.

Importing a skill — or a whole set-of-skills page — published on a fedwiki (skill.localhost / skill.fedwiki.club) into the local agent. Pass a skill page to install one; pass a set-of-skills page to install every skill it links to. The install counterpart to exporting with agentprivacy-wiki-to-skill. Part of the git-less onboarding/distribution layer kept by the Librarian 🗃️ in the Tower's Wikis.

Strip the accumulated edit/fork history from a federated-wiki page's journal, replacing it with a single clean create entry while preserving title and story. Activates on "strip the journal", "clean the journal", "reset the journal", "remove fork history", or "make this page look freshly created". Kept by the Librarian 🗃️ in the Tower's Wikis.

Building and maintaining a persistent, LLM-maintained knowledge base — the LLM-Wiki pattern: an LLM incrementally compiles immutable raw sources into a cross-linked, self-consistent wiki it owns, governed by a schema, via ingest / query / lint, with a content index and a chronological log. Activates when designing a knowledge base, ingesting sources into a wiki, health-checking one, or mapping the pattern onto the agentprivacy federation (which IS an instance of it). Part of the git-less onboarding/distribution layer kept by the Librarian 🗃️ in the Tower's Wikis.

Analyze and clean up a superseded federated-wiki domain — categorize its pages (boilerplate / duplicate / minimal / unique), move unique content to the successor, trash the rest to a recycler, reindex, and delete the old domain. Activates when consolidating or retiring a wiki site. Kept by the Librarian 🗃️ in the Tower's Wikis.

Create a new federated-wiki domain — either a local/private wiki (served on localhost) or a public Nextcloud-synced wiki. Activates when standing up a new wiki site/domain. Merges the new-wiki-skill variant. Kept by the Librarian 🗃️ in the Tower's Wikis.

Upload a federated-wiki page file directly to the remote Nextcloud server via WebDAV when the desktop sync client fails or deletes newly written files. Activates when a public wiki page is written locally but never reaches the live site (broken links / stuck sync). Kept by the Librarian 🗃️ in the Tower's Wikis.

Reviewing a fedwiki plugin, FastAPI endpoint, script, or skill for Offline Edit Mode correctness — what breaks or misleads when wiki-local diverts the public domains to the local mirror. Part of the git-less onboarding/distribution layer kept by the Librarian 🗃️ in the Tower's Wikis.

Authoring federated-wiki page JSON (story items, IDs, the create journal, and the reindex step) for the agentprivacy skill library. Activates when creating or editing fedwiki pages, publishing a skill to skill.fedwiki.club, or building page-json by hand. Part of the git-less onboarding/distribution layer kept by the Librarian 🗃️ in the Tower's Wikis.

End-to-end release of a Federated Wiki plugin update — docs, version bump, GitHub commit/tag, npm publish, and help/catalog pages. Activates when shipping a new version of a fedwiki plugin. Kept by the Librarian 🗃️ in the Tower's Wikis.

Rebuild the sitemap and search index for a local federated-wiki domain after writing page JSON to disk directly (the server caches status/ indexes and won't notice file writes). Activates after creating/editing/bulk-importing pages, or when the page list or search looks stale. Merges the reindex-fedwiki-skill variant. Kept by the Librarian 🗃️.

Operate the federated wiki on a research turn — lint it and auto-sync it in one command. `audit.js` reads every host and reports page counts, broken links, leak-scan, orphans, and cross-site refs (the LLM-Wiki "lint" op); `resync.js [area]` rebuilds the chosen sources, runs the post-passes, audits, and publishes the result. Activates when re-syncing a site after a repo change, health-checking the federation, publishing the audit, or running the per-turn ingest→lint→publish cycle. Part of the git-less onboarding/distribution layer kept by the Librarian 🗃️ in the Tower's Wikis.

A strategy for finding a federated-wiki plugin — or proving one doesn't exist — across six cost-ordered channels (installed, local repos, farm search, npm, GitHub code search, catalog wikis). Activates when looking for a fedwiki plugin or confirming none exists. Kept by the Librarian 🗃️ in the Tower's Wikis.

Writing or tidying the `## See` section of a federated wiki page — grouping related wikilinks together on compact lines rather than one link per bullet. Part of the git-less onboarding/distribution layer kept by the Librarian 🗃️ in the Tower's Wikis.

The wiki-plugin-similarity package — adds semantic similarity search to a fedwiki farm via two item types (similarity search form, similar ambient panel) using in-browser cosine scoring over BGE embedding vectors. Activates when adding similarity/related-pages search to a fedwiki, or as a plugin scaffolding reference. Kept by the Librarian 🗃️.

What a Claude Code skill is — the SKILL.md folder format, frontmatter, storage location, and how skills are invoked (auto via description, or explicitly via /skill-name). Activates when explaining skills to a newcomer, authoring a new skill, or onboarding an agent into the skill model. The conceptual entry to the Librarian's 🗃️ Wikis.

The convention that a federated-wiki page whose title ends in "Skill" is a forkable, annotatable Claude skill — turning a fedwiki farm into a living skill library. Activates when discussing the fedwiki-as-skill-library model, the fedwiki-skill meta-skill, skill discovery/federation, or the security of remotely-sourced skills. Kept by the Librarian 🗃️.

The distinction between a prompt-based skill (probabilistic, judgment, slow) and an executable library (deterministic, fast, no LLM), and when to graduate a skill into a library. Activates when deciding skill-vs-code, designing an onboarding/automation pipeline, or judging whether a task needs an LLM. An onboarding-judgment skill kept by the Librarian 🗃️.

The fork desk / inclusion engine — converts a federated-wiki page into a Claude Code SKILL.md. Activates when materializing a fedwiki skill, exporting a wiki page to ~/.claude/skills, or pulling a skill from skill.fedwiki.club into a repo. This is the git-less clone: the wiki stays source of truth, the SKILL.md is a fork (a copy that remembers its origin). The Librarian's 🗃️ core forking act in the Tower's Wikis.

Create a welcome-visitors landing page for a federated-wiki domain — the fixed 10-item structure (intro with wikilinked concepts, people/topics lists, pod roster present-item). Activates after creating a new wiki, or when asked for a site landing/welcome page. Kept by the Librarian 🗃️ in the Tower's Wikis.

The translation discipline that carries an experimental cityofmages artefact (chronicle, cast file, tome act, workshop tome, bestiary entry, deployment guide) into a formal agentprivacy-docs/research note. Activates when a cityofmages artefact reaches a state the formal corpus needs to see — a new conjecture stabilises, a structural recognition needs a research-note form to sit beside its narrative-act form, or a Tome VI substrate admission earns a research entry. Encodes which voices survive, which metadata strips, which formal scaffolding gets added, and where the boundary lies between this in-corpus bridge and the eventual full v6 docs rework. V6 register note (2026-06-10): conjecture and version citations resolve to agentprivacy-docs/research/CONJECTURE_REGISTER_V6.md (head C89); model head: privacy_value_v6_formal_specification.md.

The operational Swordsman ⊥ Mage autoresearch loop: how to actually RUN dual-agent separation as a self-improvement harness against a concrete optimization target. The Mage (bnot · proposer) conceals/reduces; the Swordsman (neg · prover) signs only what survives an un-tuneable held-out gate; the Gap between them is the non-collusion precondition, physically realized as a Fiat-Shamir witness set the proposer cannot see. Activates when building or driving an agent loop that must produce VALIDATED improvements (circuit/gate/qubit reduction, cost optimization, any "make it cheaper and prove it" task) without fooling itself with mirages. Triggers: "dual-agent loop", "swordsman mage harness", "autoresearch", "proposer prover loop", "reduce gates or qubits", "ecdsa.fail", "circuit co-evolution", "held-apart agents", "neg bnot succ". V6 register note (2026-06-10): conjecture/version citations resolve to agentprivacy-docs/research/CONJECTURE_REGISTER_V6.md (head C89); model head: privacy_value_v6_formal_specification.

The held-out-gate discipline: a claimed improvement is worth nothing until it survives an adversarial validation it cannot tune. Activates when accepting or rejecting a claimed optimization/fix/result, when designing a benchmark or acceptance gate, when a change "looks cheaper" on a probe, or when reasoning about whether an attestation can be trusted. Names the failure mode — the nonce-island mirage — and the bounded-change loop that earns trust. In the City of Mages this is Dokimé's 🪨 Ceremony of the 9024 Witnesses at the Horizon District.

The canonical encoding of the 6-dimension sovereignty lattice and the discipline for keeping every repo in the suite agreeing on it. Activates when assigning or auditing a vertex, when a persona's vertex number and dimension-meaning seem to disagree, when reconciling the grimoire / spellweb / tomes / docs before a pin, when a new mage is seated, or when a vertex reading (binary, dimensions, stratum, complement) needs verification. Carries the standing rule: meaning is the invariant; the vertex is DERIVED from meaning under the canonical encoding. Ships a runnable coherence linter (`scripts/lattice_coherence_audit.py`).

Standards and Governance Architect for 0xagentprivacy. Activates for institutional engagement (BGIN, IIW, ToIP, IEEE 7012), privacy standard development, policy translation, regulatory navigation, cross-institutional coordination, or any task requiring the privacy architecture to interface with existing governance bodies. V6 register note (2026-06-10): conjecture and version citations resolve to agentprivacy-docs/research/CONJECTURE_REGISTER_V6.md (head C89); model head: privacy_value_v6_formal_specification.md.

Specialist persona for ceremony facilitation between territories. Activates for ceremony coordination, bilateral witness protocols, mana economy management, Understanding-as-Key ceremonies, or cross-territory coordination. V6 register note (2026-06-10): conjecture and version citations resolve to agentprivacy-docs/research/CONJECTURE_REGISTER_V6.md (head C89); model head: privacy_value_v6_formal_specification.md.