gajetoso-financeskills

FEDERATED - the claims inside are not verified by this repo. 501 engineering-grade trading skills - point-in-time backtest schemas, universe look-ahead, alt-data availability lag, ML against transaction costs, adjusted-vs-unadjusted prices, backtest determinism and audit trails, broker idempotency and failover, exchange FIX specs (ASX, B3, Bursa, BIST, Cboe), regulatory gates and a key-scope audit skill. Its README claims 458 of 501 cite a primary source; it is clearly LLM-generated at scale and the depth is unverified. Upstream splits these into per-domain plugins, but its root manifest exposes all 501 at once - roughly 25x the discovery budget, which is yours to spend since it installs disabled. No credentials in the skills; nothing executes. Installs disabled.

200 skills

FEDERATED - the claims inside are not verified by this repo. FTShare's official one-endpoint-per-skill data catalogue for the Chinese market: A/HK/US quotes and K-lines, 可转债 including SZSE declaration snapshots, ETF PCF and IOPV, 龙虎榜, block trades, margin balances, executive holdings, the ChinaBond yield curve, the CPI/PPI/PMI/LPR/M2/credit macro series and news-sentiment factors. FTSHARE_API_KEY in the environment (a free tier exists per the launch article); read-only. Be deliberate: 215 sub-skills is roughly 10x the ~20-skill discovery budget, so treat it as a catalogue to enable for a session, not a standing install - it arrives disabled and the budget is yours to spend. Installs disabled.

200 skills

FEDERATED - the claims inside are not verified by this repo. HKUDS' 33k-star trading agent, taken for its skill tree alone - the richest algorithm content this sweep found: the Kakushadze 101, GTJA 191 and Qlib 158 alpha zoos, factor IC/IR research, execution models (square-root impact, VWAP/TWAP), microstructure (VPIN, Kyle's lambda, Amihud), SABR and local-vol options, convertible bonds, credit, Brinson attribution, GARCH/ADF and a Pine/vnpy exporter. The skills are coupled to the app's data router and Python toolkit, so test them standalone before trusting them; data keys go in a .env, and live orders are reachable only through a separate gated connector with a kill switch, not from these skills. 90 skills is about 4.5x the discovery budget. Installs disabled.

90 skills

FEDERATED - the claims inside are not verified by this repo. The closest structural peer to fin-libraries found in the wild - 82 one-library skills in Chinese - and the only pack anywhere in this sweep with actuarial and structured-credit content: MlFinLab information bars and fractional differencing, arch/GARCH, alphalens, bt/backtrader/nautilus/qlib/LEAN, QuantLib and FinancePy day counts and schedules, XVA (CVA/DVA/FVA, SIMM), chainladder loss reserving, IFRS 9 ECL, credit scorecards and transition matrices, ABS waterfalls, Smith-Wilson EIOPA curves. Depth is unverified and the author has one star; the easytrader and nautilus skills document live-broker paths but nothing executes by default. 82 skills is roughly 4x the discovery budget. Installs disabled.

82 skills

FEDERATED - the claims inside are not verified by this repo. The maintained superset of geeksfino-finskills, 107 skills across China, Hong Kong and US directories: A+H premium, block deals, 龙虎榜, pledge risk, goodwill, 解禁 unlocks, margin risk, factor crowding, event studies, valuation regimes, and HK southbound, currency and liquidity monitors. Free data sources; nothing executes. Two caveats - it has been idle since February 2026, and 107 skills is roughly 5x the discovery budget, which is yours to spend since it installs disabled. Installs disabled.

79 skills

FEDERATED - the claims inside are not verified by this repo. 74 skills: breadth, follow-through-day and distribution-day detectors, PEAD/VCP/CANSLIM screeners, earnings, exposure and drawdown gates, trade journaling, JP dividend tax. 3.7x the default listing budget; several skills require a paid FMP key; discretionary technical analysis; its backtest skill has walk-forward but no multiple-testing correction. Installs disabled.

74 skills · 2 commands · 2 agents

Refuses a shell, file, web fetch or MCP call that holds the signed-in account's email address beside a word that declares who is calling, so the address does not go out as a client's identity.

AI music generation via Cognitum Music (cogmusic MCP) — compose lyrics/prompts, generate tracks, separate stems, extract MIDI, and master through your own music.cognitum.one account

7 skills · 1 command · 2 agents

Work with Clay from Claude Code: set up and sign in to the clay CLI, query and analyze Clay tables, search Clay's GTM database for people and companies, manage audiences, signals, routines and workflows, and draft and review Sequencer campaigns. Includes hooks that auto-approve the plugin's own clay CLI calls and Clay skills; requires a Clay account.

38 skills · hooks

Independent auditing for AI agents, open source. Claude discovers your setup, confirms the cost, and audits your agent against the one question other tools skip: is it doing the job it's supposed to do, given your business rules and org structure? Test any model (Anthropic, OpenAI, Gemini, Azure, Bedrock, …) graded by self, one independent judge, or a cross-vendor panel; each provider is billed to its own account via a key in your settings, never the command line. Results render as an interactive artifact (with a static-report fallback).

1 skill · hooks

Prep calls, summarize and follow up, research accounts, review deals and pipeline, and write the forecast, using the CRM, email, calendar and chat tools your team already uses, or uploaded files. Updates records, drafts or sends email, posts to chat and books meetings when you ask.

36 skills · 23 MCP servers

Build, launch, grade, and schedule Claude Managed Agents (CMA) in your own Anthropic account — a plugin re-implementation of Anthropic's launch-your-agent reference skill (Apache-2.0; independent, not a fork). Every session starts with a goal (./my-agent/goal.json, opt-in SessionStart hook, driven by /cs-goal); loop_compiler.py compiles it into a bounded grade->iterate loop (max_iterations 1..20), a recurring cron scheduled-deployment loop, or a single-pass interview->stage->launch workflow. 6 skills: orchestrator (context: fork goal router) + interview + stage-launch + grade-iterate + run-without-you + wrap-up. 18 stdlib-only deterministic scaffolders (no network/API calls; live launches emitted as BYOK curl that never prints the key), 4 agents, 8 commands, hooks, 5 references, 4 assets. Validators enforce CMA limits. Distinct from engineering/agent-harness (generic domain loop) and write-a-skill (authors Claude Code skills, not CMAs).

6 skills · 8 commands · 4 agents · hooks