Skip to content
CAI
Software that uses CAICheck a score

anthropics/financial-services

59.2

Weak · 15 November 2026

2.3k

lines of production code

Python

with JavaScript

5

measurements over time

CAI band scale
CAI trend line
CAI lens gauges

What this system is

Features

New admin plugin for configuring Claude Office add-in cloud access

The \claude-for-msft-365-install\ plugin is introduced to help administrators provision direct cloud access (Vertex AI, Bedrock, Azure AI Foundry, or LLM gateway) for the Claude Office add-in. It provides CLI commands to generate customized add-in manifests, handle Azure admin consent, and write per-user configuration via Microsoft Graph extension attributes. The plugin also includes scripts to export user chat history and skills from local browser storage, and to clear add-in caches on macOS and Windows to resolve stale configuration issues.

claude-for-msft-365-install · high confidence

New financial services agent plugins for earnings, reconciliation, KYC, and research

This change introduces a suite of new agent plugins under \plugins/agent-plugins/\, each providing a specialized workflow for financial services tasks. The \earnings-reviewer\ agent automates post-earnings updates by pulling transcripts and filings to update models and draft notes. The \gl-reconciler\ handles daily GL-to-subledger reconciliations, identifying breaks and tracing root causes. The \kyc-screener\ parses onboarding documents, runs KYC/AML rules, and screens against sanctions lists. The \market-researcher\ generates sector primers, competitive landscapes, and peer comps. Supporting these are shared skills like \audit-xls\ for spreadsheet integrity checks, \xlsx-author\ and \pptx-author\ for headless file generation, and specific domain skills like \break-trace\ and \competitive-analysis\. Each plugin includes a \plugin.json\ manifest and an \agents/\*.md\ file defining the agent's role, tools, and guardrails.

plugins/vertical-plugins · high confidence

New repository validation, deployment, and orchestration scripts

Added a suite of scripts to manage and verify managed-agent cookbooks and plugins. \scripts/check.py\ validates YAML/JSON syntax, frontmatter, and file references across plugins and managed agents, while also ensuring git hooks are installed. \scripts/deploy-managed-agent.sh\ automates the deployment of managed agents by resolving manifest conveniences (inlining system files, uploading skills, creating subagents) and posting to the Anthropic API. \scripts/orchestrate.py\ provides a reference event loop for cross-agent handoffs with strict allowlisting and schema validation. \scripts/sync-agent-skills.py\ re-syncs bundled agent skills from vertical-plugin sources. \scripts/test-cookbooks.sh\ runs dry-run deployments to assert well-formed agent bodies. \scripts/validate.py\ performs harness-side schema validation for worker output. \scripts/version\_bump.py\ enforces plugin version bumps on changes via pre-commit hooks and CI checks.

scripts · high confidence

Behavioural changes

2 commits (1 fix) modifying plugins/agent-plugins

A change to existing behaviour in plugins/agent-plugins — 2 commits (1 fix), 92 files: “Restructure repo and add named agents (\#81)”; “fix(plugins): add SKILL.md frontmatter, repair hooks.json, dedupe spglobal MCP (\#243)” (refs \#81, \#243).

plugins/agent-plugins · medium confidence · unverified

Automated plugin version bumping on commit

A new pre-commit hook has been added to automatically patch-bump any plugin that has staged changes, ensuring the plugin version is exactly one patch level ahead of the main branch. This automation runs the \scripts/version\_bump.py\ script during the commit process, provided Python 3 is available, and can be bypassed for individual commits using the \--no-verify\ flag.

.githooks · high confidence

Relocate and refine number extraction script for pitch agent

The \extract\_numbers.py\ script, previously located in the financial-analysis plugin, has been moved to the pitch-agent plugin under \plugins/agent-plugins/pitch-agent/skills/ib-check-deck/scripts/\. This change also corrects a logic error in the \normalize\_number\ function: it now iterates over unit keys sorted by length in descending order to ensure longer unit strings (like 'billion') are matched before shorter ones (like 'bn'), preventing incorrect normalization of financial figures.

python · high confidence

Dependencies

Added Python dependency manifests for example and plugin skills

New requirements.txt files have been introduced to define Python dependencies for specific components. The python-bootstrap example now requires fastapi, uvicorn, and PyJWT\[crypto\]. The DCF Model Builder skill, available in both the pitch-agent and model-builder plugin paths, specifies openpyxl (\>=3.0.0) for Excel handling and requests (\>=2.28.0) for HTTP operations. Notably, the model-builder version of the DCF model has removed the yfinance dependency in favor of the generic requests library.

(dependencies) · high confidence

Written by cca.canine.dev from the codebase's own history, inside the signed delivery this page is composed from.

How this codebase got here

Score

  • CAI 59 → 59 (+0.1)
  • Rubric changed (rubric-2026.11.1 → rubric-2026.11.4) — scores are not directly comparable.

Lenses

  • Code Health 89 → 82 (-7.4)
  • Architecture 99 → 99 (+0.0)
  • Maturity 54 → 59 (+4.5)
  • Readiness 41 → 41 (+0.0)
  • Security 94 → 94 (+0.0)
  • Performance 100 → 100 (+0.0)

Resolved (7)

  • Duplicated block (11 lines × 2) (plugins/agent-plugins/pitch-agent/skills/ib-check-deck/scripts/extract_numbers.py)
  • Duplicated block (5 lines × 2) (scripts/check.py)
  • Hotspot: claude-for-msft-365-install/scripts/build-manifest.mjs (claude-for-msft-365-install/scripts/build-manifest.mjs)
  • No tests found
  • Off-boarding risk: anonymized user #1
  • check.check_refs (cognitive 28) (scripts/check.py)
  • check.check_refs (cyclomatic 19) (scripts/check.py)

New (1)

  • Off-boarding risk: anonymized user #1

Changes since last survey

  • 1 commits — 0 feature/other, 1 fixes

By area

  • claude-for-msft-365-install/commands — 1 commit

Notable commits

  • fix: Msft 365 install fixes (#403)

Architecture

  • Unchanged — 0 containers · 1 contexts · 0 edges

Written by cca.canine.dev from the codebase's own history, inside the signed delivery this page is composed from.

Survey your own repository

anthropics/financial-services was measured the same way every project in this corpus was: the same rubric, at a pinned commit, with the result published in full. Point a surveyor at a repository you know and see whether you agree with it.

About this page

  • The score is its most recent published measurement, taken on 15 November 2026 at a pinned commit. It is not a live figure and does not change until the project is measured again.
  • Measured at commit 3e32c2287dcb637c184c01cc0ad82d6af7e9bb0f — the exact code this score is about.
  • Scored under rubric-2026.11.4 — the same rubric and the same method as every other entry in this index.
  • Measured by Canine Code Analyzer (cca.canine.dev) using cca measure-2026.11.73+preprod-52de90b74216.