Interactive worksheet for the collaborative pruning session · built July 16 · usage measured over the last 45 days. Nothing has been deleted — your taps here are the decisions.
How this works: every row has response buttons. A dashed outline marks my recommended default — tap it to accept, or pick differently. Rows with no dash are ones that genuinely need your eye (mostly zero-use skills). GSD and plugin packs are decided at the group level (amber rows). Add a note to any row with ＋ note. Your choices save on this device as you go. When you're done (or partway), hit Copy response and paste the block back to me in chat or a Dub note.
Your skills — the biv-skills repo (~/.claude/commands)
64 skills · 22 invoked in the last 45 days · 14 carry audit findings. These are yours — the packs below are installed third-party. Finding #38 additionally applies namespace-wide: 42 of these files still carry obsolete name: frontmatter (invocation is by path, so it's dead weight) — one sweep commit fixes all of them.
Remaining unset in this section →
C01ops:briefing#29 FIX
~/.claude/commands/ops/briefing.md
Daily morning briefing — intelligent daily digest written as markdown to ~/Reports/briefing/; audio is on-demand via gen-report-tts.sh
Stale Google Workspace MCP tool names — actively used (33 invocations), so this fix matters.
C02utilities:dub#35 FIX
~/.claude/commands/utilities/dub.md
Emit an audio-worthy deliverable as a Dub spoken-document report, or parse a pasted DUB-FEEDBACK block and act on its notes
Slug-reuse guidance contradicts the sent-Dubs-are-immutable rule — amend Job 1.
C03utilities:cb#34 FIX
~/.claude/commands/utilities/cb.md
Copy to Clipboard — copy code blocks or text to system clipboard via pbcopy with clean formatting
'Offer proactively' section teaches the exact anti-pattern biv.md bans — rewrite to auto-copy-same-turn.
C04clients:health-check#33 FIX
~/.claude/commands/clients/health-check.md
Full business health analysis — pipeline, revenue, client relationships, prospects, growth trajectory. Spawns specialist agents for parallel analysis, synthesizes into actionable b
Stale Gmail tool names (#29) + hardcodes an Airtable base/table ID — should resolve from generated bindings/schema.js.
C05clients:audit-contacts#29 FIX
~/.claude/commands/clients/audit-contacts.md
Parallel contact audit — spawns 10 subagents to clean, enrich, and link 50 contacts at a time using Gmail, Airtable, and reasoning
References Google Workspace MCP tool names that no longer exist; its batch-metadata step has NO modern equivalent — needs a redesign, not a rename.
C06clients:sweep#29 FIX
~/.claude/commands/clients/sweep.md
Client data-quality sweep — audit Clients/People/Deals for missing, stale, and orphaned data; queue fixes to ClickUp
Stale Google Workspace MCP tool names (get_gmail_* → readGmail*).
C07utilities:meeting-log#39 FIX
~/.claude/commands/utilities/meeting-log.md
Log a MacWhisper transcription — or any audio/video file — to the Obsidian vault as a diarized note with AI-generated summary and action items as checkboxes in the transcript note.
Needs the documented Obsidian-race fallback in Step 4.
C08biv:ship-and-verify#28 QUESTION
~/.claude/commands/biv/ship-and-verify.md
Stage → commit → push → wait for GH Action deploy → fetch deployed URL through CF Access → assert key substring. One-shot deploy loop with verification.
Is the deploy-helper flow still live? If retired: retire-sweep needed (wiki Claude Code Skills page cites it).
C09clients:pipeline#31 RETIRE?
~/.claude/commands/clients/pipeline.md
Pipeline overview — show all Clients grouped by stage with key metrics
Audit read: retire (prospect pipeline paused; script it drives is a candidate too).
C10clients:prospects:triage#32 FIX
~/.claude/commands/clients/prospects/triage.md
Surface 5-10 qualified prospects weekly with agent reasoning about fit, post summary to Discord, create OmniFocus review task
Description doesn't match stdout behavior; Discord-identity leftover; discovery paused so empty triage is expected.
Is the biweekly sprint ritual still live? Also carries dead memory refs + stale Gmail tool names (#29).
C12pitch:event-press-release#37 FIX
~/.claude/commands/pitch/event-press-release.md
Draft a music event press release (festival, party series, tour) in the BiV / INFAMOUS PR house style synthesized from ~20 reference releases (HARD Summer, Elements, CRSSD, Splash
References a reference corpus + ~/Scripts/google-drive/ implementation that no longer exist.
C13reporting:social-metrics#36 FIX
~/.claude/commands/reporting/social-metrics.md
Manual social campaign metric refresh via a Drafts form. For social-seeding campaigns backed by Airtable CCC+CR records and a per-campaign phase2-section generator. Bradley ships m
Calls nonexistent MCP tool create_records (plural); stale commit trailer.
C14utilities:beatport-isrc#40 FIX
~/.claude/commands/utilities/beatport-isrc.md
Pull ISRC(s) from a Beatport release URL and write to the matching Airtable Campaign(s)
Hardcoded Airtable table ID — repoint to generated bindings + live describe_table check.
C15biv:fleet-wave0 uses / 45d
~/.claude/commands/biv/fleet-wave.md
Run a multi-agent build wave on an iOS/web repo — compose a LOCKED file-disjoint unit roster, dispatch Opus implementers in ephemeral worktrees, Sonnet adversarial verifiers fix in
C16clients:audit-orgs0 uses / 45d
~/.claude/commands/clients/audit-orgs.md
Parallel org audit — spawns subagents to validate, enrich, and clean Clients table
C17clients:deal-intake0 uses / 45d
~/.claude/commands/clients/deal-intake.md
Ingest an inbound deal from an email thread — extract entities, dedup against the CRM, and create the Client + Person + Deal records
C18clients:followup0 uses / 45d
~/.claude/commands/clients/followup.md
Draft a follow-up email for a client or prospect based on relationship context, pipeline stage, and outreach history
C19clients:intake0 uses / 45d
~/.claude/commands/clients/intake.md
New lead intake — scan Gmail for inbound inquiries, match against existing records, surface new prospects for review
C20clients:onboard0 uses / 45d
~/.claude/commands/clients/onboard.md
New client onboarding — create all records, folders, and tracking across Airtable, OmniFocus, GDrive, and Obsidian
C21clients:prep0 uses / 45d
~/.claude/commands/clients/prep.md
Pre-call brief — pull client/org data, recent meeting notes, open action items, and campaign performance into a concise brief
C22clients:prospects:dossier0 uses / 45d
~/.claude/commands/clients/prospects/dossier.md
Compile a research dossier for a prospect label -- why they surfaced, roster, releases, genre fit, conversation starters
Client performance report — campaign results, placement stats, and ROI summary for a specific client or org
C25gsd:dev-preferences0 uses / 45d
~/.claude/commands/gsd/dev-preferences.md
Load developer preferences into this session
C26kickoff:ingest-assets0 uses / 45d
~/.claude/commands/kickoff/ingest-assets.md
Ingest campaign assets a client sent by email (attachments + metadata) into GDrive and the Campaign record, so the Fillout form pre-fills and they don't re-enter what they already
Capture a substantive synthesis into the knowledge layer. BiV-relevant syntheses become wiki pages (canonical) and are picked up by the RAG index on sync; non-BiV personal insights
C34rag:ingest0 uses / 45d
~/.claude/commands/rag/ingest.md
Add content to the personal Karpathy RAG database. The bundled version handles manual notes (title + content + tags). Extend it with custom pipelines for your own sources (session
C35rag:init0 uses / 45d
~/.claude/commands/rag/init.md
Initialize (or verify) the personal Karpathy RAG database at ~/.rag-db/rag.db. Creates the SQLite database with FTS5 from schema.sql. Trigger when the user says "init the RAG", "bo
C36rag:status0 uses / 45d
~/.claude/commands/rag/status.md
Quick dashboard for the personal Karpathy RAG database. Shows counts, document types, recent activity, source coverage. Trigger when the user says "RAG status", "how big is the kno
C37reporting:bef0 uses / 45d
~/.claude/commands/reporting/bef.md
Process Batch Email Feedback — fetch curator emails, interpret with agent reasoning, write Campaign Results to Airtable
C38reporting:log-outcome0 uses / 45d
~/.claude/commands/reporting/log-outcome.md
Manually log a non-DSP campaign outcome (radio airplay / social post / press hit / pass) → writes a Campaign Results row with deterministic Placement linking. Mirrors BEF v3 archit
C39reporting:songstats0 uses / 45d
~/.claude/commands/reporting/songstats.md
Run Songstats correlation pipeline — fetch API data, correlate with Airtable Curator/Placement/Campaign Results records, trigger platform enrichment
C40reporting:submithub0 uses / 45d
~/.claude/commands/reporting/submithub.md
Auto-export SubmitHub CSV + ingest — one command does everything
Pre-trim SubmitHub curator comments — read each raw comment, produce a cleaned client-facing summary + confidence/disposition, flag low-confidence items for Bradley. BEF-style clau
C42research:contact-hunt0 uses / 45d
~/.claude/commands/research/contact-hunt.md
Headless contact-hunt — name-search + URL-chase a contact-less curator and emit a structured, evidence-bearing candidates list, EMAIL-first. FINDS a contact only; it NEVER composes
C43utilities:conventions0 uses / 45d
~/.claude/commands/utilities/conventions.md
One-line summary of what the skill does
C44utilities:decide0 uses / 45d
~/.claude/commands/utilities/decide.md
Decision cockpit — generate a single-file interactive HTML tool to grapple with many decisions at once, then parse the compiled answer pasted back into chat
C45utilities:dub-digest0 uses / 45d
~/.claude/commands/utilities/dub-digest.md
Turn one or more source inputs (video/podcast/thread/article URLs or files) into analyzed Dub report threads — one Dub per input, lensed against a project's live context
C46utilities:obsidian0 uses / 45d
~/.claude/commands/utilities/obsidian.md
Operate Bradley's Obsidian vault (BBVault01) — write new notes, append/edit existing notes, set properties, and know where things belong. Routes to the right folder with house-conv
C47wiki:fix0 uses / 45d
~/.claude/commands/wiki/fix.md
LLM repair pass for wiki-lint HARD failures the deterministic linter can't autofix — resolves the high-confidence semantic cases (link retargets, dead paths, index entries), commit
C48wiki:lint0 uses / 45d
~/.claude/commands/wiki/lint.md
Lint the BiV wiki — validate wikilinks, detect banned conventions, surface orphans. Auto-fixes unambiguous cases, appends findings to wiki/log.md.
C49wiki:reconcile0 uses / 45d
~/.claude/commands/wiki/reconcile.md
Fold Bradley's edits in the "Workspace Wiki - BB Notes" vault back into the prod wiki. Isolates his changes via the vault's git baseline (NOT a vault-vs-prod diff), checks each pag
C50ops:cup-capture34× / 45d
~/.claude/commands/ops/cup-capture.md
Capture a ticket to ClickUp with Definition of Ready enforcement. Usage: /cup-capture <text> [--bug] [--decide] [--priority high] [--group <name>] [-d description]
C51ops:ai-digest28× / 45d
~/.claude/commands/ops/ai-digest.md
Daily AI learning digest — fetches curated sources, synthesizes a markdown digest, files ClickUp tickets for crisp workspace triggers; audio is on-demand via gen-report-tts.sh
C52rag:query25× / 45d
~/.claude/commands/rag/query.md
Search and synthesize from the RAG index — the FTS5 search layer over the BiV wiki, raw corpus, and personal documents. Uses full-text search, tag/type filtering, and wikilink rela
C53utilities:tts25× / 45d
~/.claude/commands/utilities/tts.md
Generate conversational TTS audio briefing from current context — writes script, sends to ElevenLabs/OpenAI, saves to iCloud
C54utilities:nice-wan24× / 45d
~/.claude/commands/utilities/nice-wan.md
Bradley's "nice wan" sign-off protocol — one-liner, housekeeping, memory scan. Triggered when Bradley says "nice wan" or "nice one".
C55wiki:fold24× / 45d
~/.claude/commands/wiki/fold.md
Fold one or more facts into the BiV wiki — amend the existing page(s) a fact touches, or draft a new atomic page when none covers it, then open one PR (no chat approval — the PR is
Run the daily reporting pipeline — Songstats → BEF → SubmitHub export + ingest. Calls scripts directly for speed.
C64wiki:deep-audit1× / 45d
~/.claude/commands/wiki/deep-audit.md
Unified wiki accuracy audit — deterministic ref check + LLM claim verification against the live system + semantic-judgment drift + (monthly) full-corpus multi-agent ground-truth pa
Loose installs (~/.claude/skills)
Mostly Cloudflare doc-mirror packs installed during past builds, plus a few flat-file duplicates of your real skills. Everything marked RETIRE? is reinstallable from its public distribution in one command — deleting these loses nothing permanent.
Remaining unset in this section →
S01agents-sdk#51 RETIRE?
~/.claude/skills/agents-sdk
Build AI agents on Cloudflare Workers using the Agents SDK. Load when creating stateful agents, durable workflows, real-time WebSocket apps, scheduled tasks, MCP servers, chat appl
Unused CF pack; reinstallable if an Agents SDK build starts.
S02cloudflare-one#45 RETIRE?
~/.claude/skills/cloudflare-one
Guides Cloudflare One Zero Trust and SASE work across Access, Gateway, WARP, Tunnel, Cloudflare WAN, DLP, CASB, device posture, and identity. Use when designing, configuring, troub
Zero-touch Cloudflare ZT pack; reinstallable from CF distribution.
S03cloudflare-one-migrations#46 RETIRE?
~/.claude/skills/cloudflare-one-migrations
Plans migrations from Zscaler ZIA/ZPA, Palo Alto, legacy VPN, SWG, or SASE stacks to Cloudflare One. Use for migration assessments, policy mapping, rollout plans, and parity/gap an
Companion to cloudflare-one; reinstallable.
S04durable-objects#52 RETIRE?
~/.claude/skills/durable-objects
Create and review Cloudflare Durable Objects. Use when building stateful coordination (chat rooms, multiplayer games, booking systems), implementing RPC methods, SQLite storage, al
Unused CF pack; reinstall if a stateful-coordination build appears.
S05sandbox-sdk#47 RETIRE?
~/.claude/skills/sandbox-sdk
Build sandboxed applications for secure code execution. Load when building AI code execution, code interpreters, CI/CD systems, interactive dev environments, or executing untrusted
Unused CF pack; reinstallable.
S06turnstile-spin#50 RETIRE?
~/.claude/skills/turnstile-spin
Set up Cloudflare Turnstile end-to-end in a project — scan the codebase, create the widget via the Cloudflare API, deploy the managed siteverify Worker, write the frontend snippets
Mirrors public CF docs; trivially recoverable the day a form needs bot protection.
S07web-perf#49 BROKEN
~/.claude/skills/web-perf
Analyzes web performance using Chrome DevTools MCP. Measures Core Web Vitals (LCP, INP, CLS) and supplementary metrics (FCP, TBT, Speed Index), identifies render-blocking resources
Requires the chrome-devtools MCP, which is not installed — cannot run today.
S08workflow-automation#48 RETIRE?
~/.claude/skills/workflow-automation
Workflow automation is the infrastructure that makes AI agents reliable. Without durable execution, a network hiccup during a 10-step payment flow means lost money and angry custom
n8n-mcp is the surviving owner of n8n work.
S09.git0 uses / 45d
~/.claude/skills/.git
S10canvas0 uses / 45d
~/.claude/skills/canvas.md
Generate interactive visual canvases in Obsidian's JSON Canvas format.
S11clean-my-ai-harness0 uses / 45d
~/.claude/skills/clean-my-ai-harness
Map the setup Claude can see and prepare a safe cleanup plan. Use after model changes or when old instructions, project files, memories, or skills may conflict.
S12clipboard0 uses / 45d
~/.claude/skills/clipboard.md
Copy Claude's last code block or specified content to the system clipboard with clean formatting. Use when user says /clipboard, /cb, or "clipboard that".
Create and script actions for the Drafts app. Use this skill when the user needs help creating custom Drafts actions, understanding action steps, using template tags, scripting wit
S15figma-swiftui-co-design0 uses / 45d
~/.claude/skills/figma-swiftui-co-design
Establish and run a collaborative Figma-to-SwiftUI workflow where a user and Codex co-edit a Figma Design file, approve exact frames for implementation, validate against real iOS r
S16tts0 uses / 45d
~/.claude/skills/tts.md
Convert text or file content to audio. Primary: ElevenLabs. Fallback: OpenAI.
S17workers-best-practices0 uses / 45d
~/.claude/skills/workers-best-practices
Reviews and authors Cloudflare Workers code against production best practices. Load when writing new Workers, reviewing Worker code, configuring wrangler.jsonc, or checking for com
S18clickup69× / 45d
~/.claude/skills/clickup
Use when managing ClickUp tasks, sprints, or comments via the `cup` CLI tool. Triggers: task queries, status updates, sprint tracking, creating subtasks, posting comments, threaded
S19brainstorm9× / 45d
~/.claude/skills/brainstorm
A dialectic thinking partner for working an idea through — developing it, reframing a problem, pressure-testing a plan, mapping a space, or deciding between options. This skill sho
S20skill-creator4× / 45d
~/.claude/skills/skill-creator
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specializ
S21graphify2× / 45d
~/.claude/skills/graphify
Use for any question about a codebase, its architecture, file relationships, or project content — especially when graphify-out/ exists, where the question should be treated as a gr
S22wrangler2× / 45d
~/.claude/skills/wrangler
Cloudflare Workers CLI for deploying, developing, and managing Workers, KV, R2, D1, Vectorize, Hyperdrive, Workers AI, Containers, Queues, Workflows, Pipelines, and Secrets Store.
S23agent-reach1× / 45d
~/.claude/skills/agent-reach
>
S24cloudflare-email-service1× / 45d
~/.claude/skills/cloudflare-email-service
Send and receive transactional emails with Cloudflare Email Service (Email Sending + Email Routing). Use when building email sending (Workers binding or REST API), email routing, A
S25rsvp1× / 45d
~/.claude/skills/rsvp
Send articles, notes, books, and RSS subscriptions to Bradley's RSVP Nano e-reader (open-source ESP32 speed-reader). Use whenever the request is to put something on the RSVP device
GSD suite (68 dirs)
Only 8 of 68 invoked in 45 days. Installer-managed — hand-deleting individual dirs does NOT work (gsd-update resurrects them; finding #53), so the rows below are evidence, not per-row choices. One decision covers the suite: surface-minimal hides the ~55 unused clusters from every session's skill list, retire entirely is decide ticket 86baz2vmp.
GSDGSD suite — one decision for all 68
G01gsd-add-tests0 uses / 45d
~/.claude/skills/gsd-add-tests
Generate tests for a completed phase based on UAT criteria and implementation
G02gsd-ai-integration-phase0 uses / 45d
~/.claude/skills/gsd-ai-integration-phase
Generate an AI-SPEC.md design contract for phases that involve building AI systems.
Create detailed phase plan (PLAN.md) with verification loop
G62gsd-quick5× / 45d
~/.claude/skills/gsd-quick
Execute a quick task with GSD guarantees (atomic commits, state tracking) but skip optional agents
G63gsd-execute-phase4× / 45d
~/.claude/skills/gsd-execute-phase
Execute all plans in a phase with wave-based parallelization
G64gsd-cleanup2× / 45d
~/.claude/skills/gsd-cleanup
Archive accumulated phase directories from completed milestones
G65gsd-complete-milestone2× / 45d
~/.claude/skills/gsd-complete-milestone
Archive completed milestone and prepare for next version
G66gsd-debug2× / 45d
~/.claude/skills/gsd-debug
Systematic debugging with persistent state across context resets
G67gsd-ingest-docs2× / 45d
~/.claude/skills/gsd-ingest-docs
Bootstrap or merge a .planning/ setup from existing ADRs, PRDs, SPECs, and docs in a repo.
G68gsd-map-codebase1× / 45d
~/.claude/skills/gsd-map-codebase
Analyze codebase with parallel mapper agents to produce .planning/codebase/ documents
Plugin packs (claude-plugins-official)
A pack's skill descriptions load into EVERY session (~35KB across the whole catalog) — a never-invoked pack is pure context tax. The decision unit is the PACK (uninstalling removes all its rows at once); individual rows below are evidence. context7 was uninstalled today per your note.
Audit and improve CLAUDE.md files in repositories. Use when user asks to check, audit, update, improve, or fix CLAUDE.md files. Scans for all CLAUDE.md files, evaluates quality aga
Manage Discord channel access — approve pairings, edit allowlists, set DM/group policy. Use when the user asks to pair, approve someone, check who's allowed, or change policy for t
Set up the Discord channel — save the bot token and review access policy. Use when the user pastes a Discord bot token, asks to configure Discord, asks "how do I set this up" or "w
Creates and maintains Figma Code Connect template files that map Figma components to code snippets. Use when the user mentions Code Connect, Figma component mapping, design-to-code
Creates and maintains Figma Code Connect template files that map Figma components to code snippets. Use when the user mentions Code Connect, Figma component mapping, design-to-code
Creates and maintains Figma Code Connect template files that map Figma components to code snippets. Use when the user mentions Code Connect, Figma component mapping, design-to-code
Creates and maintains Figma Code Connect template files that map Figma components to code snippets. Use when the user mentions Code Connect, Figma component mapping, design-to-code
**MANDATORY prerequisite** — you MUST invoke this skill BEFORE every `create_new_file` tool call. NEVER call `create_new_file` directly without loading this skill first. Trigger wh
**MANDATORY prerequisite** — you MUST invoke this skill BEFORE every `create_new_file` tool call. NEVER call `create_new_file` directly without loading this skill first. Trigger wh
**MANDATORY prerequisite** — you MUST invoke this skill BEFORE every `create_new_file` tool call. NEVER call `create_new_file` directly without loading this skill first. Trigger wh
**MANDATORY prerequisite** — you MUST invoke this skill BEFORE every `create_new_file` tool call. NEVER call `create_new_file` directly without loading this skill first. Trigger wh
Use this skill when implementing a Figma design as code (design → code) — the read-FROM-Figma direction. Triggers: 'implement this Figma design', 'build this screen from Figma', 't
Use this skill when implementing a Figma design as code (design → code) — the read-FROM-Figma direction. Triggers: 'implement this Figma design', 'build this screen from Figma', 't
**MANDATORY prerequisite** — you MUST invoke this skill BEFORE calling the `get_design_context` Figma MCP tool. You MUST trigger this skill whenever the user wants to implement, bu
Use this skill alongside figma-use when the task involves translating an application page, view, or multi-section layout into Figma. Triggers: 'write to Figma', 'create in Figma fr
Use this skill alongside figma-use when the task involves translating an application page, view, or multi-section layout into Figma. Triggers: 'write to Figma', 'create in Figma fr
Use this skill alongside figma-use when the task involves translating an application page, view, or multi-section layout into Figma. Triggers: 'write to Figma', 'create in Figma fr
Use this skill alongside figma-use when the task involves translating an application page, view, or multi-section layout into Figma. Triggers: 'write to Figma', 'create in Figma fr
MANDATORY prerequisite — load this skill BEFORE every `generate_diagram` tool call. NEVER call `generate_diagram` directly without loading this skill first. Trigger whenever the us
MANDATORY prerequisite — load this skill BEFORE every `generate_diagram` tool call. NEVER call `generate_diagram` directly without loading this skill first. Trigger whenever the us
MANDATORY prerequisite — load this skill BEFORE every `generate_diagram` tool call. NEVER call `generate_diagram` directly without loading this skill first. Trigger whenever the us
MANDATORY prerequisite — load this skill BEFORE every `generate_diagram` tool call. NEVER call `generate_diagram` directly without loading this skill first. Trigger whenever the us
Build or update a professional-grade design system in Figma from a codebase. Use when the user wants to create variables/tokens, build component libraries, create individual compon
Build or update a professional-grade design system in Figma from a codebase. Use when the user wants to create variables/tokens, build component libraries, create individual compon
Build or update a professional-grade design system in Figma from a codebase. Use when the user wants to create variables/tokens, build component libraries, create individual compon
Build or update a professional-grade design system in Figma from a codebase. Use when the user wants to create variables/tokens, build component libraries, create individual compon
Translates Figma motion and animations into production-ready application code. Use when implementing animation/motion from a Figma design — user mentions "implement this motion", "
Translates Figma motion and animations into production-ready application code. Use when implementing animation/motion from a Figma design — user mentions "implement this motion", "
Translates Figma motion and animations into production-ready application code. Use when implementing animation/motion from a Figma design — user mentions "implement this motion", "
Translates Figma motion and animations into production-ready application code. Use when implementing animation/motion from a Figma design — user mentions "implement this motion", "
SwiftUI ↔ Figma translation. Use whenever the user mentions Swift, SwiftUI, iOS, iPhone, or iPad — in EITHER direction — translating a Figma design into SwiftUI (design → code), or
SwiftUI ↔ Figma translation. Use whenever the user mentions Swift, SwiftUI, iOS, iPhone, or iPad — in EITHER direction — translating a Figma design into SwiftUI (design → code), or
SwiftUI ↔ Figma translation. Use whenever the user mentions Swift, SwiftUI, iOS, iPhone, or iPad — in EITHER direction — translating a Figma design into SwiftUI (design → code), or
SwiftUI ↔ Figma translation. Use whenever the user mentions Swift, SwiftUI, iOS, iPhone, or iPad — in EITHER direction — translating a Figma design into SwiftUI (design → code), or
This skill helps agents use Figma's use_figma MCP tool in the FigJam context. Can be used alongside figma-use which has foundational context for using the use_figma tool.
This skill helps agents use Figma's use_figma MCP tool in the FigJam context. Can be used alongside figma-use which has foundational context for using the use_figma tool.
This skill helps agents use Figma's use_figma MCP tool in the FigJam context. Can be used alongside figma-use which has foundational context for using the use_figma tool.
This skill helps agents use Figma's use_figma MCP tool in the FigJam context. Can be used alongside figma-use which has foundational context for using the use_figma tool.
This skill helps agents use Figma's use_figma MCP tool in the Slides context. Can be used alongside figma-use which has foundational context for using the use_figma tool.
This skill helps agents use Figma's use_figma MCP tool in the Slides context. Can be used alongside figma-use which has foundational context for using the use_figma tool.
This skill helps agents use Figma's use_figma MCP tool in the Slides context. Can be used alongside figma-use which has foundational context for using the use_figma tool.
This skill helps agents use Figma's use_figma MCP tool in the Slides context. Can be used alongside figma-use which has foundational context for using the use_figma tool.
**MANDATORY prerequisite** — you MUST invoke this skill BEFORE every `use_figma` tool call. NEVER call `use_figma` directly without loading this skill first. Skipping it causes com
**MANDATORY prerequisite** — you MUST invoke this skill BEFORE every `use_figma` tool call. NEVER call `use_figma` directly without loading this skill first. Skipping it causes com
**MANDATORY prerequisite** — you MUST invoke this skill BEFORE every `use_figma` tool call. NEVER call `use_figma` directly without loading this skill first. Skipping it causes com
**MANDATORY prerequisite** — you MUST invoke this skill BEFORE every `use_figma` tool call. NEVER call `use_figma` directly without loading this skill first. Skipping it causes com
Guidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't rea
Guidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't rea
Guidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't rea
Guidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't rea
Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verificati
Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verificati
Use when starting any conversation - establishes how to find and use skills, requiring skill invocation before ANY response including clarifying questions
Use when starting any conversation - establishes how to find and use skills, requiring skill invocation before ANY response including clarifying questions
Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any succ
Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any succ
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design befo
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design befo
Use when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for m
Use when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for m
Use when starting feature work that needs isolation from current workspace or before executing implementation plans - ensures an isolated workspace exists via native tools or git w
Use when starting feature work that needs isolation from current workspace or before executing implementation plans - ensures an isolated workspace exists via native tools or git w
Open forks — the rest of the session agenda
Not skills — the decisions you said you want to work through together. Go = do it as described, Hold = park it, Discuss = session item.
F1Hooks: NotchBar AgentStatus (#25)
claude-event-hook.sh is wired into all 8 hook event blocks but Bartender's own plist shows the last hook received June 24 — dead three weeks. Retiring means also disabling the agentStatus provider in Bartender's NotchBar settings (it installed these entries itself).
F2Empty agents dir (#27)
~/.claude/agents/_retired/ is empty and untracked — rmdir, zero risk. Held only because you asked to review pruning together.
F3Git rule SSoT (#8)
The global CLAUDE.md git paragraph and biv.md's git section duplicate each other. Which owns it — wire biv.md into biv repos as its header claims, or keep the global copy and shrink biv.md?
F4Main-push strength (#17)
Global says NEVER push to main; biv.md says ask-first. Absolute or per-instance-authorizable? Pick one wording.
F5Voice config keys (#16)
settings.json carries both "voice": {enabled, mode} and top-level "voiceEnabled" — needs one /config toggle to see which key the current build writes, then delete the other. Two minutes, interactive.
F6biv.md's reason to exist (your s5 note)
Domain buckets (musicpromotion.md, consulting.md, development.md) vs the current one-manual shape. Feeds the D6 routing-card design — held so the session decides the shape before anything is drafted.
F7The approval principle (86baz2vq5)
Your candidate: you gate what needs your domain judgment, your voice, or client-facing surfaces — everything mechanical is the agents'. Distill it, then collapse the five enforcement layers to match.
F8Context architecture (86baz0ah7)
Memory vs wiki vs ClickUp vs .planning vs live code vs ephemeral artifacts — the first-principles consolidation you asked for. Shape-linter is off the table per your note.
Overall comment — anything that doesn't fit a row: