uygnoey/skills-from-claude-blog/finance-narrative-integrity-layer
Claude Code and agent skill by uygnoey/skills-from-claude-blog
- community
Checking that the SKILL.md file is still in the repository.
Keep finance narratives (board decks, monthly reviews) coherent as numbers refresh by using Claude as an integrity layer for reconciliation, consistency checks, and first-pass variance commentary.
Install
npx vskill@latest install uygnoey/skills-from-claude-blog/finance-narrative-integrity-layerSKILL.md
The same SKILL.md was published earlier as uygnoey/skills-from-claude-blog/finance-narrative-integrity-layer.
Purpose
Use Claude to maintain a single coherent story behind financial numbers as inputs change, by (1) reconciling narrative claims to a source of truth, (2) reviewing the narrative like a board reader, and (3) drafting recurring variance commentary in an established voice.
When to use
- You ship a board deck where numbers refresh up to the last minute and multiple stakeholders edit in parallel.
- You publish a monthly financial review that must keep consistent voice and conventions month over month.
- You inherit or open a model you don’t fully understand and want an initial driver/structure sanity check.
Inputs
- A board deck or narrative doc (slides or document).
- A single source of truth for the numbers (model output tables, dashboards, or approved extracts).
- Prior-month commentary (for voice/convention consistency).
Instructions
1) Reconcile claims to a single source of truth
- Provide Claude the narrative document/deck and the latest source-of-truth tables.
- Use the reconciliation prompt template: templates/reconcile-claims.md
- Apply fixes in the narrative and rerun whenever numbers move.
2) Read the narrative like a board member
- Use the board-reader prompt template: templates/board-reader-review.md
- Resolve contradictions, missing context, and places where the narrative assumes internal knowledge.
3) Draft first-pass variance commentary in an established voice
- Provide the month’s table and prior-month commentary.
- Use the variance draft template: templates/variance-commentary-first-pass.md
- Edit the output for judgment, nuance, and stakeholder alignment.
4) Separate projects by audience (memory hygiene)
- Keep separate Claude projects for distinct audiences (e.g., board deck vs. monthly review) so tone and conventions remain stable.
- Commit important documents to project memory.
5) Extract decisions from long cross-functional threads
- When a long thread produces a key conclusion, ask Claude to extract the conclusion and reasoning and store it where you will reuse it.
Examples
- Board deck workflow (checklist): examples/board-deck-checklist.md
- Monthly review workflow (checklist): examples/monthly-review-checklist.md
- New model first-look prompts: examples/new-model-first-look.md
Source
This skill is derived from: https://claude.com/blog/how-anthropics-finance-team-uses-claude-to-shape-the-narrative-behind-the-numbers
Security
- Security scanPassed, 95 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
Popularity
- Repository stars
- 4
- Repository forks
- 1
- 7-day trend score
- 13
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/uygnoey/skills-from-claude-blog/finance-narrative-integrity-layer)