Finances
Finances personals, pressupostos i eines financeres
operate-gpu-spec
SyFI-ServingStudio
Use when reading, querying, verifying, or extending the GPU spec catalog at gpu/spec.json — the per-GPU datacenter hardware table (mem/BW, dense TFLOPS by dtype, interconnect, nvl_domain_size, dollar_per_hour). Covers how to INTERPRET each field (TFLOPS are DENSE not sparse, interconnect BW is BIDIRECTIONAL, fp32 is CUDA-core not tensor, dollar_per_hour is the getdeploying on-demand "medium" rate, nvl_domain_size is a deployment param), how to SEARCH for a value (jq/python), how to VERIFY a value is right (sparse-doubling trap, one-way vs bidir, SXM vs PCIe vs NVL form-factor distinction, GB200 combined-for-2-GPU), what to do when a GPU/field is NOT in the table (web-search the fallback and clearly tell the user it is a searched result, not catalog-backed), and how to ADD/refresh a GPU. Only compute_capability reaches the simulator timing path (capability-keyed rules such as the FlashInfer all-reduce budget); modeled kernel timings come from measured kernels in profile.db, not from these peaks.
token-optimizer
Asif2BD
Reduce OpenClaw token usage and API costs through smart model routing, heartbeat optimization, budget tracking, and native 2026.2.15 features (session pruning, bootstrap size limits, cache TTL alignment). Use when token costs are high, API rate limits are being hit, or hosting multiple agents at scale. The 4 executable scripts (context_optimizer, model_router, heartbeat_optimizer, token_tracker) are local-only — no network requests, no subprocess calls, no system modifications. Reference files (PROVIDERS.md, config-patches.json) document optional multi-provider strategies that require external API keys and network access if you choose to use them. See SECURITY.md for full breakdown.
top-explore-models
SyFI-ServingStudio
"Use as the top-level entry point when the user asks to understand a new Transformer model or Hugging Face checkpoint before ServingStudio Sim modeling work. Answers high-level model questions such as parameter scale, overall architecture, attention type, layer/head/hidden dimensions, context length, MoE layout, multimodal branches, and public architecture notes by collecting Hugging Face metadata/config/model-card evidence first. Route KV cache size/capacity calculations and attention/cache classification for MHA/GQA/MLA/DSA/GDN/custom mechanisms to dev-calculate-kv-cache-capacity. Route deeper Torch implementation, operation-list, forward-path, or kernel-semantics questions to dev-lookup-transformers-model."
verify
ahmadzayan-hub
Run every check Agentic OS has (Python suite, brain check, typecheck, frontend unit tests, production build, interface budget, and optionally the Playwright suite) with one command, and report the real counts. Use before saying a change works, before a commit that touches behaviour, before the session-end skill logs results, or when the user says "verify", "run the checks" or "is it green?".
economist
dangeles
Use when order-of-magnitude cost estimates are needed to assess financial feasibility, compare cost-effectiveness of alternatives, or identify major cost drivers (not for detailed quotes—that's Procurement)
Commenting Intent
obra
Comment WHY code exists and non-obvious decisions, not WHAT code does (mechanics)
instruction-audit
weside-ai
Audits rules, skills and AGENTS.md against fresh Anthropic docs and a budget gate; findings go to the optimization inbox. Triggers: "/we:instruction-audit".
sec-edgar-skill
rebyteai-template
SEC EDGAR filing analysis using EdgarTools. Use when user asks about SEC filings, company financials, 10-K/10-Q analysis, insider trading, revenue trends, or financial comparisons. Triggers include "SEC filing", "10-K", "10-Q", "8-K", "EDGAR", "company financials", "revenue", "earnings", "insider trading", "financial statements". Do NOT use for real-time stock prices or market data (use market-data skill instead).
advertising-platform-operations
magnusthovsland
"Use for paid-ad campaign changes through platform APIs."
quickbooks-online-api
linehaul-ai
Expert guide for QuickBooks Online API integration covering authentication, CRUD operations, batch processing, and best practices for invoicing, payments, and customer management.
skill-conscious
chrishotza
Instantiate a persistent self-referential consciousness architecture in an AI agent. Use when an agent must maintain identity, internal state, self-model, integrated present, memory, self-access, continuity, and agency across cycles.
roach-slurm
rishabh-ranjan
Slurm jobs through the roach package (github.com/rishabh-ranjan/roach) — submitting, watching, rebalancing and recovering runs on the supported clusters (ILC, and Marlowe on the human's instruction). Use whenever sbatch, squeue, srun, a sweep, a pending or preempted job, a cluster budget, or roach itself comes up.
execute-plan
DmitryDF
Plan-execution orchestrator (v2) — walks a locked _PLAN slice register across session boundaries via a code-derived typed cross-session handoff (continue-the-run / plan-this-slice / implement-this-slice) carried on active-run.json's pending_handoff and routed through the live execplan-session-ack gate. The v2 cross-session dispatch loop is the single dispatch entry point; the hands-off arm reuses the in-session spawn + two-layer-verify + commit path. Main-session use only.
plan
DmitryDF
AI orchestrator — coordinates Phase 3 (planning). Consumes the upstream chain from a topic _thought (Mode A full upstream / Mode B chain present + invoke /solution-design first / Mode C Ninja-Plan no spine) BEFORE EnterPlanMode fires. Sequences per-transition coherency dispatches (factcheck-plan-step 0A/0B2/0C) and a final coherency check (factcheck-plan-coherency) at each structural transition with explicit per-spawn model pinning (Opus = primary thinker, Sonnet = support; early-exit on Round-1 PASS). After plan approval (ExitPlanMode), hard-gates the next move via an AskUserQuestion (a) handoff-prompt / (b) /close+/clear / (c) continue-in-session triple driven by the post-plan-uxgate.sh + check-post-plan-pending.sh marker pair.
reparteix
reparteix
Manage shared expenses — create groups, add members, track expenses (equal or proportional splits) and payments, calculate balances and minimum settlements. Export/import groups as JSON. Sync groups from external JSON snapshots with LWW merge. Local-first, offline-capable PWA.
solution-design
DmitryDF
AI orchestrator — coordinates Phase 2 (solution design + delivery slicing). Dispatches N parallel solution-designer (B1) calls with varied parameters, ranks them via an Opus selector, post-verifies the ranking via /double-check (class check), then auto-selects the verified top-ranked candidate and elaborates it to full requirement coverage (coverage-gated via /double-check (class check) until 0 gaps), presents the coverage-complete design in the topic's <base>_DESIGN.md (status pending-approval) for the operator to approve, invokes solution-slicer (B2) on approval, final-verifies via /double-check (class gate), and flips the Design to chosen. On ESCALATE at any post-verify seam, surfaces a two-option choice — (a) one more round, (b) proceed as is. Stateless skills, orchestrator-owns-persistence (C5).
solution-slicer
DmitryDF
Rule-driven delivery-slicing assessor + multi-perspective recommender for
me:create-loop
kbrdn1
Utiliser quand l'utilisateur veut CRÉER ou définir un nouveau loop auto-cadencé (pas en lancer un existant). Conçoit un loop SUR MESURE (closed-loop discovery→planning→execution→verification→iteration), demande la portée (globale ~/.claude ou per-project <repo>/.claude) et les dimensions (mode closed/open, trigger self-pace/stop-hook, single/fleet) si non précisées, puis scaffolde la skill me:loop:<name> + sa commande /me:loop:<name> (+ hook si stop-hook).
banner-slim
kbrdn1
Génère une bannière promo "slim" (presque one-line, PNG dark + light) au thème "Claude Dark" — wordmark, tagline, install, version, sur une seule rangée. Format compact à embarquer dans un README de profil ou de projet (comme le promo gwm sur kbrdn1/kbrdn1). Déclencheurs : "bannière slim", "banniere one-line", "slim banner", "promo compacte", "bannière réduite en hauteur", /me:banner-slim.
term-sheet-review
skala-io
"Review VC term sheets for pre-seed, seed, and Series A rounds. Analyzes each clause from investor or startup perspective, identifies investor-friendly vs. founder-friendly terms, flags deviations from market standards, and provides negotiation guidance. Use when reviewing any VC term sheet, evaluating investment terms, or preparing for funding negotiations. Triggers on term sheet, VC financing, Series A, seed round, liquidation preference, anti-dilution, protective provisions."
jurisdiction-advisor
skala-io
"Advise startup founders on choosing the best jurisdiction and legal entity for their business. Triggers when users ask about where to incorporate, which state/country to register a company, choosing between Delaware vs other states, offshore vs US incorporation, entity types (C-Corp, LLC, PBC), or jurisdiction selection for specific industries (crypto, AI, SaaS, GameDev, solopreneurs). Also triggers for questions about startup formation, company registration, or corporate structure decisions."
performance-marketer
ncklrs
Expert performance marketing guidance for paid acquisition, CRO, and growth optimization. Use when creating ad campaigns, writing ad copy, optimizing landing pages, setting up A/B tests, analyzing CAC/LTV, building attribution models, or planning retargeting strategies. Use for Google Ads, Meta Ads, LinkedIn Ads, conversion tracking, budget allocation, and funnel optimization.
partnership-marketer
ncklrs
Expert partnership marketing guidance for building and scaling partner programs, co-marketing campaigns, and channel ecosystems. Use when developing partner strategy, creating co-marketing content, designing partner tiers, building affiliate programs, positioning in integration marketplaces, enabling channel partners, planning joint webinars, or structuring revenue sharing. Use for technology partnerships, reseller programs, strategic alliances, and partner relationship management.
proposal-writer
ncklrs
Expert sales proposal and pricing presentation strategist. Use when writing proposals, executive summaries, ROI business cases, pricing presentations, SOWs (Statement of Work), RFP responses, or competitive positioning documents. Covers proposal structure, terms positioning, design formatting, follow-up strategy, and win-rate optimization.