What does this skill do, and when should you use it?
This skill supports fixed-income relative-value analysis by assessing whether a bond is rich, cheap, or fairly valued versus comparable instruments and relevant curves. It uses MCP tools to obtain bond pricing, risk-free curves, and credit curves for spread decomposition. It also runs parallel rate-shock scenarios and can add historical pricing context. The intended output is a spread decomposition, scenario P&L table, and explicit rich/cheap conclusion.
Calls bond_price to obtain clean and dirty price, yield, duration, convexity, DV01, and Z-spread; uses interest_rate_curve in a list-then-calculate workflow to interpolate government or swap curves and compute G-spread; uses credit_curve to search and calculate credit spreads by country and issuer type; calculates residual spread after the credit component; runs -100bp, -50bp, 0, +50bp, and +100bp parallel-rate scenarios with yieldbook_scenario; optionally calls tscc_historical_pricing_summaries and fixed_income_risk_analytics; synthesizes the results into spread decomposition, scenario P&L, and buy, overweight, avoid, underweight, or neutral recommendations.
- A fixed-income analyst compares bonds from the same issuer or comparable issuers for relative richness or cheapness.
- A portfolio manager evaluates whether a bond's current spread compensates for its rate, credit, and residual risks.
- A trader decomposes total spread into risk-free, credit, and liquidity or technical components.
- A risk team estimates price changes and P&L per 100 notional under parallel rate shocks.
- A researcher adds historical spread context and Z-score analysis to a current valuation view.
How do you install this skill?
- Core functionality depends on multiple LSEG MCP tools whose schemas and availability are not supplied; confirm subscription, permissions, regional reachability, and data as-of times before use.
- The skill requests a definitive rich/cheap recommendation but does not define handling for missing data, tool failures, model differences, or unsupported bond types.
- Treat outputs as analyst work product requiring qualified fixed-income review, not as standalone investment advice or trading instructions.
- MCP Server
bond_priceinterest_rate_curvecredit_curveyieldbook_scenariotscc_historical_pricing_summariesfixed_income_risk_analytics
In Cowork, open Settings → Plugins → Add plugin, paste https://github.com/anthropics/financial-services, and select the relevant vertical plugin from the marketplace list. Alternatively, upload a repository zip containing plugins/partner-built/lseg/. The source does not document a standalone installation command for this individual skill.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/anthropics/financial-services.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/plugins/partner-built/lseg/skills/bond-relative-value" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- Analyze the relative value of the bond with ISIN [ISIN]. Calculate its G-spread, credit-curve spread, and residual spread, then run -100bp, -50bp, 0, +50bp, and +100bp rate scenarios.
With the relevant MCP tools configured, submit a request such as: “Analyze the relative value of the bond with ISIN [ISIN]. Calculate its G-spread, credit-curve spread, and residual spread, then run -100bp, -50bp, 0, +50bp, and +100bp rate scenarios.” Add one or more bond identifiers for comparison. LSEG MCP access may require a provider subscription or API key.
What are this skill's strengths and limitations?
- Covers pricing, yield curves, credit curves, and rate-scenario analysis.
- Explicitly requires decomposition into total, credit, and residual spread components.
- Supports bond comparisons and optional historical and deeper risk analytics.
- Defines structured outputs for spread decomposition, scenario P&L, and the valuation conclusion.
- Depends on multiple MCP tools; the core workflow cannot be completed without suitable data connections.
- The source provides no standalone test suite or validation results for this skill.
- The scenario workflow focuses on parallel rate shifts and does not document non-parallel curve shocks.
- It does not provide investment, legal, tax, or accounting advice; outputs require qualified professional review.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Bond Relative Value Analyst this page ✓ Anthropic · Official | 41 · Not recommended | ★ 40k | 1mo ago | Apache-2.0 |
| Fixed-Income Portfolio Review ✓ Anthropic · Official | 44 · Not recommended | ★ 40k | 1mo ago | Apache-2.0 |
| Bond Futures Basis Analyzer ✓ Anthropic · Official | 46 · Use with care | ★ 40k | 1mo ago | Apache-2.0 |
| Swap Curve Strategy Analyst ✓ Anthropic · Official | 42 · Not recommended | ★ 40k | 1mo ago | Apache-2.0 |
| Three-Statement Model Builder ✓ Anthropic · Official | 55 · Use with care | ★ 40k | 1mo ago | Apache-2.0 |
How did FollowSkills review this skill?
The skill names bounded analytical tools and shows a data flow from pricing and curves to synthesized output, but it does not specify least privilege, user confirmation, sensitive financial-data handling, external effects, rollback, or full disclosure of provider data flows. The absence of malware, credential theft, covert exfiltration, or destructive defaults avoids a red-line deduction, but these governance gaps materially reduce the score.
The tool chain and five-step workflow are broadly self-consistent, including stated two-phase calls for some tools. However, there are no parameter or response schemas, availability checks, abnormal-input handling, controlled failure messages, or committed reproducibility tests. Static calibration caps the score at 10, and these omissions justify further deduction.
The name, description, and bond-relative-value use cases are reasonably clear, covering rich/cheap analysis, spread decomposition, and rate shocks. Inputs, non-fit cases, precise semantic triggers, missing-data behavior, Chinese-language support, and mainland-China network reachability are not defined, so invocation and environment fit remain uncertain.
The document has a readable structure with a tool inventory, workflow, and output templates. Repository-level evidence supplies an Apache-2.0 license, verified official-organization provenance, and general maintenance/CI signals. The skill itself lacks installation and dependency notes, versioning, changelog, examples, FAQs, known limitations, and explicit maintenance ownership, preventing a high score.
The workflow addresses pricing, risk-free curves, credit curves, scenarios, historical context, and provides directly reusable table formats. Nevertheless, the file does not verify the underlying tool capabilities or calculations, and it gives limited guidance on complete inputs, missing data, recommendation thresholds, or alternatives. Results therefore still require substantial expert review, warranting a below-mid-to-upper score.
The named steps, tools, and output fields provide some auditability. Repository CI validates plugins and scans for secrets, but the supplied files do not show skill-specific tests or third-party execution evidence covering key paths. This is a low-confidence static conclusion with limited corroboration, so the score remains below the static ceiling.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →