← 返回 Skills 市场
henrijoe20

Corporate Credit Memo

作者 henrijoe20 · GitHub ↗ · v1.0.0
cross-platform ✓ 安全检测通过
309
总下载
1
收藏
1
当前安装
1
版本数
在 OpenClaw 中安装
/install corporate-credit-memo
功能描述
Generates institutional-grade corporate credit application memoranda (Credit Memos) in English from uploaded annual reports, financial statements, or user-pr...
使用说明 (SKILL.md)

Corporate Credit Memo Skill

Produces full institutional-grade Credit Application Memoranda (20–30+ pages) from uploaded financial documents and user-provided deal parameters. Output follows the 8-section structure used by major international banks, with all financial ratios calculated and flagged where data is insufficient.


Step 1 — Intake & Clarification

When triggered, immediately check what the user has provided. Do NOT ask multiple questions at once. Follow this sequence:

1a. Check for uploaded files

If annual reports / financial statements are attached → proceed to Step 2. If nothing is attached, ask:

"Please upload the company's annual reports (last 2–3 years, PDF or Word). If you don't have them, paste the key financials and I'll work with those."

1b. Collect deal parameters (ask once, in a single message)

Once files are confirmed, ask the user to confirm or provide:

1. Borrower legal name
2. Facility type & amount (e.g. GBP 50m term loan / USD 120m revolving credit)
3. Tenor (e.g. 3 years, 5 years)
4. Purpose of loan (working capital / capex / refinancing / acquisition)
5. Security / guarantee structure (if known)
6. Repayment profile (bullet / amortising / quarterly)
7. Pricing (if known — e.g. SONIA + 250bps)
8. Lending bank / branch name
9. Any internal policy or industry classification to reference?
   (If unsure, leave blank — I will note it as [TO BE CONFIRMED])

Tell the user:

"Items marked [TO BE CONFIRMED] will be flagged in the report for your team to complete — including RAROC, internal limits, and any policy-specific thresholds."


Step 2 — Document Extraction

Read all uploaded files carefully. Extract and organise:

From financial statements (P&L, Balance Sheet, Cash Flow):

  • 3 years of annual data + latest interim period if available
  • All line items needed for ratio calculation (see references/financial-ratios.md)
  • Auditor name and opinion type (unqualified / qualified / emphasis of matter)
  • Reporting currency and any FX translation notes

From the annual report narrative:

  • Business description, segment breakdown, revenue mix
  • Key customers and suppliers (concentration risk)
  • Management team and ownership structure
  • Geographic footprint
  • Material litigation, contingent liabilities, off-balance-sheet items
  • ESG disclosures if present

Flag clearly any data that is:

  • Missing → label [DATA NOT AVAILABLE — please provide]
  • Restated vs prior year → note the restatement
  • Unaudited → label [UNAUDITED]

Step 3 — Research (Web Search)

Use web search to supplement document data for:

  • Industry overview: market size, growth rate, cycle positioning
  • Regulatory / policy environment relevant to the borrower's sector
  • Competitive landscape: 2–3 named peers with approximate market share
  • Recent news on the company (last 12 months): M&A, rating actions, management changes
  • Macroeconomic context relevant to the jurisdiction and sector

Do NOT fabricate specific figures. If web search returns insufficient data, note:

"[Industry data limited — recommend analyst review / Bloomberg supplement]"


Step 4 — Calculate Financial Ratios

Read references/financial-ratios.md for all formulas and flag logic.

Calculate every ratio for each available year. Apply the following flag rules:

  • ✅ Calculated successfully → show figure
  • ⚠️ Calculated but at borderline threshold → show figure + note
  • 🔲 Data insufficient → show [N/A — data required]
  • 📋 Requires internal input → show [TO BE CONFIRMED — internal policy]

RAROC and internal capital allocation are always [TO BE CONFIRMED]. DSCR denominator must match the proposed repayment schedule — confirm with user if unclear.


Step 5 — Draft the Credit Memo

Read references/credit-memo-structure.md for the full 8-section structure and content requirements for each section.

Drafting principles:

  • Write in formal institutional English — consistent with major UK/international bank standards
  • Tone: analytical, direct, no marketing language
  • Every material statement should be traceable to either the uploaded documents or web search
  • Clearly distinguish facts from analytical judgements
  • Use tables for all financial data — never prose for numbers
  • Section headers must match the structure in references/credit-memo-structure.md exactly
  • Flag all [TO BE CONFIRMED] items inline — do not omit them or leave blank
  • Do not invent internal policy figures (hurdle rates, limits, RWA) — always flag these

Risk section (Section 6):

Read references/risk-framework.md before drafting. Assign a risk rating (Low / Medium / High) for each risk category with a one-line rationale. This section reflects the analytical approach expected of a senior risk function, consistent with PRA supervisory expectations for credit risk governance.

Disclaimer:

Append the standard disclaimer from assets/disclaimer.md at the end of every report.


Step 6 — Output as Word Document (.docx)

Generate the final report as a formatted .docx file using the docx skill conventions:

  • Font: Arial throughout
  • Colour scheme: dark navy headers (#1F4E79), mid-blue subheadings (#2E75B6)
  • All financial tables: dark navy header row, alternating white/light-blue rows
  • Info tables (deal terms, company details): two-column label/value format
  • Risk rating boxes: colour-coded text (green=Low, amber=Medium, red=High)
  • Page header: "Credit Application Memorandum | Confidential — Internal Use Only"
  • Page footer: document date + page number
  • Cover page: borrower name, facility type, amount, date, classification label

Present the .docx file to the user when complete. Also offer a brief verbal summary of the key credit highlights and top 3 risks.


Quality Checklist (run before presenting output)

Before finalising, verify:

  • All 3 years of financials extracted and populated in tables
  • Every ratio in references/financial-ratios.md either calculated or flagged
  • No internal policy figures invented (RAROC, RWA, limits)
  • Risk section has a rating for each of the 7 categories
  • All [TO BE CONFIRMED] items visible and not hidden
  • Disclaimer appended
  • Document is .docx format, not markdown

Handling Edge Cases

Listed company: Extract financials from uploaded annual report. Use web search for stock performance context and peer comparison. Note exchange and ticker.

Unlisted / private company: Flag absence of public ratings. Rely on audited accounts. Note that valuation of pledged equity (if any) requires independent appraisal.

Holding company borrower / SPV: Distinguish between holdco and opco financials. Analyse guarantor separately. Note structural subordination risk.

Insufficient financials (only 1 year available): Complete what is possible. Note that trend analysis is limited and recommend additional data before credit approval.

Chinese-language source documents: Extract data from tables regardless of language. For narrative sections, translate key disclosures as needed. Note source language.

M&A / acquisition finance: Produce pro-forma combined financials. Separate target and acquirer analysis. Reference references/credit-memo-structure.md Section 1 addendum for transaction analysis structure.

安全使用建议
This skill appears internally consistent and appropriate for generating credit memos, but before installing or using it consider: (1) confidentiality — uploaded annual reports/financials can contain sensitive or regulated data, so confirm where files are processed/stored (LLM provider logs, third-party services) and that this meets your institution's data-handling policy; (2) validation — many items are flagged [TO BE CONFIRMED] (RAROC, RWA, internal policy thresholds) and must be completed by your internal teams before submission to a credit committee; (3) parsing capability — the skill claims to handle PDFs, Word, and some non-English documents but relies on the agent/platform for extraction/translation; test with non-sensitive examples to confirm accuracy and formatting (DOCX layout, tables, colour scheme); (4) regulatory/legal checks — the skill is an aid, not legal or regulatory advice; ensure legal and compliance review of security/enforceability sections before relying on them. If you need higher assurance, ask the author (or repo) where file extraction and web queries are executed (locally/on-platform/third-party) and whether any telemetry or logs are retained.
功能分析
Type: OpenClaw Skill Name: corporate-credit-memo Version: 1.0.0 The corporate-credit-memo skill is a professional financial analysis tool designed to generate structured credit reports from uploaded documents. It follows a rigorous 8-section workflow (defined in SKILL.md and references/credit-memo-structure.md), calculates 28 financial ratios (references/financial-ratios.md), and performs targeted web research for industry context. The skill demonstrates good security posture by explicitly instructing the agent not to fabricate sensitive internal bank data (like RAROC or RWA) and by appending a comprehensive disclaimer (assets/disclaimer.md). No indicators of data exfiltration, malicious execution, or prompt injection were found.
能力评估
Purpose & Capability
Name/description (generate credit memos from uploaded reports) aligns with the actual requirements and behavior: no environment variables, no unrelated binaries, and included references support the described output. The claim to handle Chinese-language source documents is plausible for an instruction-only skill but implicitly depends on the agent or platform having appropriate parsing/translation capabilities.
Instruction Scope
SKILL.md stays within the stated purpose: it directs intake questions, reading of uploaded financial documents, local reference files, and read-only web search to supplement analysis. It explicitly flags internal items as [TO BE CONFIRMED] and forbids fabrication. The instructions do involve processing potentially sensitive uploaded documents — appropriate confidentiality controls should be considered on the hosting platform, but this is expected for the skill's function.
Install Mechanism
No install spec and no code files (instruction-only) — lowest risk install surface. Nothing is downloaded or written to disk by an installer according to the manifest.
Credentials
No environment variables, credentials, or config paths are requested. The skill does not require unrelated secrets or system access; requested inputs (uploaded reports, deal parameters) are proportionate to the task.
Persistence & Privilege
always is false and the skill does not request persistent system-wide privileges or to modify other skills. Autonomous invocation is allowed (platform default) but not combined with other elevated privileges here.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install corporate-credit-memo
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /corporate-credit-memo 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
Initial release - institutional grade credit memo generator
元数据
Slug corporate-credit-memo
版本 1.0.0
许可证
累计安装 1
当前安装数 1
历史版本数 1
常见问题

Corporate Credit Memo 是什么?

Generates institutional-grade corporate credit application memoranda (Credit Memos) in English from uploaded annual reports, financial statements, or user-pr... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 309 次。

如何安装 Corporate Credit Memo?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install corporate-credit-memo」即可一键安装,无需额外配置。

Corporate Credit Memo 是免费的吗?

是的,Corporate Credit Memo 完全免费(开源免费),可自由下载、安装和使用。

Corporate Credit Memo 支持哪些平台?

Corporate Credit Memo 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Corporate Credit Memo?

由 henrijoe20(@henrijoe20)开发并维护,当前版本 v1.0.0。

💬 留言讨论