FX Carry Trade Analysis
Evaluates FX carry trades by combining spot rates, forward points, volatility surfaces and one year of spot history. It computes carry-to-vol ratios by tenor and produces a recommendation with direction, risks and conviction.
Evaluates FX carry trades by combining spot rates, forward points, volatility surfaces and one year of spot history. It computes carry-to-vol ratios by tenor and produces a recommendation with direction, risks and conviction.
This Skill guides an assistant through a carry trade assessment for a currency pair. It pulls the spot rate, prices forwards, maps the forward curve, reads the implied volatility surface and checks historical spot trends. The assistant then calculates annualized carry at each tenor and divides it by ATM implied vol to get a carry-to-vol ratio. Use it when you want to compare tenors for a pair, judge whether the carry is worth the spot risk, or see how skew and tail risk look before entering a trade. The Skill relies on MCP tools for pricing data (spot, forward, forward curve, vol surface, historical pricing and interest rate curves). The assistant does the risk-adjusted calculations and the write-up. You get a carry profile table for 1M, 3M, 6M and 1Y, a vol surface summary table (ATM vol, 25-delta put and call, risk reversal, butterfly), and a recommendation per trade. Each recommendation lists pair and direction, tenor, annualized carry, carry-to-vol ratio, skew signal, key risks and a conviction level of high, medium or low, along with position sizing guidance. Source: https://github.com/anthropics/financial-services/tree/574ed3624aebd0418c7e96cd101262f30210ab26/plugins/partner-built/lseg/skills/fx-carry-trade. Published on Finch under the upstream Apache-2.0 license; the license file ships with the package.
ipfs://bafkreifz3h2xwsba25zmz3ppnbi76f7m2ajpaiudkjonsd5byxd37cgvlm
v1 · Initial release on Finch, from anthropics/financial-services at commit 574ed36.
The primary-document hash is the SHA-256 of the plaintext primary Markdown file. The content-commitment hash covers the verified SkillRoot structure, or plaintext ZIP for older Skills; it is not the hash of the downloaded file.
Opaque identifiers and fingerprints are hidden from the primary experience but remain available for exact verification.
0xef999638f8a04b3bdba126a151e6767c37861c4797bac56b7df3884ac785cc4b0x7def8d63533bd8ec8c7959d1aa2b0f5b39ed5011b77fcbf846478706fdecb597