Vendor Anthropic financial skills: data-free (21) + data-backed wired to Rebyte data (14)#3
Open
zeroxaa wants to merge 2 commits into
Open
Vendor Anthropic financial skills: data-free (21) + data-backed wired to Rebyte data (14)#3zeroxaa wants to merge 2 commits into
zeroxaa wants to merge 2 commits into
Conversation
Import 21 skills from anthropics/financial-services (Apache 2.0, commit 4aa51ed) that carry no external-data dependency — they run purely on user-supplied inputs, logic, or formatting. They sit as top-level siblings to the repo's own data/ and backtesting/ skills. - Output/check engines: xlsx-author, pptx-author, audit-xls, clean-data-xls, ib-check-deck, deck-refresh - Model math: lbo-model, merger-model, returns-analysis, unit-economics - Doc drafting/process: ic-memo, teaser, process-letter, cim-builder, dd-checklist, dd-meeting-prep, deal-screening, deal-tracker, investment-proposal, kyc-doc-parse, kyc-rules Excluded meta-skills (skill authoring, not finance): skill-creator, ppt-template-creator. Supplied runtime assets the source referenced but never shipped: - lbo-model/scripts/recalc.py (replaces /mnt/skills/public/xlsx/recalc.py) - lbo-model/examples/LBO_Model.xlsx (standard LBO template skeleton) Add LICENSE.anthropic-skills (Apache 2.0) and ATTRIBUTION.md. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Import 14 skills from anthropics/financial-services (Apache 2.0, commit 4aa51ed) whose external-data dependency our stack can serve. Each SKILL.md gets an appended "Data sources (Rebyte)" section mapping its workflow steps to the financial/sql lake (us.fundamentals, us.eod, us.bars_1m, us.news + cn.* tables), financial/search semantic news search, and the live REST proxies (stocks = Polygon US, cn-stocks = Tushare CN). Inputs we cannot power (analyst consensus, earnings calendar, 13F/ownership, M&A deal databases, TAM estimates) are explicitly marked user-supplied in each wiring section. - Equity research: earnings-analysis, idea-generation, initiating-coverage, model-update, morning-note, sector-overview, thesis-tracker - Financial analysis: comps-analysis, dcf-model, 3-statement-model, competitive-analysis - Investment banking: strip-profile, datapack-builder, pitch-deck Excluded after assessment (core input is data we lack): catalyst-calendar (forward event calendar), earnings-preview (consensus + earnings date + implied move), buyer-list (PE fund DB, 13F, M&A transaction DB). Local fixes: dcf-model recalc.py dependency satisfied by copying lbo-model/scripts/recalc.py (references repointed); dangling upstream example references in comps-analysis / strip-profile rewritten as not bundled. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This branch stacks on
vendor-anthropic-data-free-skills(previously pushed, unmerged), so the PR delivers both batches fromanthropics/financial-services(Apache 2.0, pinned commit4aa51ed).Commit 1 — data-free skills (21)
Skills with no external-data dependency, vendored as-is: output/check engines (xlsx-author, pptx-author, audit-xls, clean-data-xls, ib-check-deck, deck-refresh), model math (lbo-model, merger-model, returns-analysis, unit-economics), document drafting/process (ic-memo, teaser, process-letter, cim-builder, dd-checklist, dd-meeting-prep, deal-screening, deal-tracker, investment-proposal, kyc-doc-parse, kyc-rules). Two upstream-referenced-but-never-shipped assets supplied locally (lbo-model recalc script + template).
Commit 2 — data-backed skills (14), wired to our data
Each of the 17 upstream data-dependent candidates was assessed against our actual catalog (financial/sql lake: us.fundamentals / us.eod / us.bars_1m / us.news + cn.* Tushare tables; financial/search semantic news; live proxies stocks=Polygon, cn-stocks=Tushare). The 14 whose core workflow our data powers are vendored, each with an appended "Data sources (Rebyte)" section in SKILL.md that maps workflow steps to concrete tables/operations and explicitly marks unpowerable inputs (analyst consensus, earnings calendar, 13F/ownership, M&A deal DBs, TAM) as user-supplied:
Excluded (core input is data we lack): catalyst-calendar, earnings-preview, buyer-list.
Local fixes in commit 2: dcf-model's cross-skill
recalc.pydependency satisfied by copying lbo-model's script and repointing references; two dangling upstream example-file references (comps-analysis, strip-profile) rewritten to state no example is bundled. ATTRIBUTION.md documents both batches, the wiring convention, and every exclusion with its reason.🤖 Generated with Claude Code