What does this skill do, and when should you use it?
This skill transfers source data into existing investment banking pitch deck templates. It covers template inspection, content mapping, formatting, table and chart handling, and iterative validation. The workflow includes backing up the original template and checking structural and visual issues. It is not intended to create presentations from scratch.
Reads relevant data from Excel, CSV, PDF, Word, database, or web sources; creates a backup of the original PowerPoint template; analyzes slide structure and maps data to template sections; populates text, bullets, actual table objects, charts, and PowerPoint shapes; standardizes units, currencies, fonts, alignment, and layout; converts presentations to PDF with LibreOffice and to images with pdftoppm for validation; fixes detected issues and repeats the validation cycle.
- An investment banking team has a branded PowerPoint pitch template and needs Excel or CSV data inserted into specified slides.
- An analyst needs to map market sizing, comparable-company, or financial-summary data from source files into an existing banking template.
- A team needs to edit already populated slides while extracting and revalidating their current content.
- A deliverable needs checks for real PowerPoint tables, text contrast, content overflow, and consistency of figures across slides.
How do you install this skill?
- The skill may handle sensitive investment-banking data but does not define redaction, access control, retention, or approval for external sources.
- Direct XML editing, file replacement, and PowerPoint rendering differences may cause corruption or visual defects; backups and human review in Microsoft PowerPoint are necessary.
- Web and other external sources are allowed without a source allowlist, required citation format, or mandatory escalation rule for conflicting data.
- No real test artifacts cover the skill's key paths, so static review cannot confirm final deck deliverability.
- Shell / CLI
- Local filesystem
LibreOfficepdftoppm
The skill is located at plugins/agent-plugins/pitch-agent/skills/pitch-deck/ and is bundled with the Pitch Agent plugin. The README supports these installation commands: claude plugin marketplace add anthropics/financial-services; claude plugin install pitch-agent@claude-for-financial-services. The supplied source does not document a standalone installation method for this skill.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/anthropics/financial-services.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/plugins/agent-plugins/pitch-agent/skills/pitch-deck" ~/.claude/skills/
rm -rf "$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:
- Populate this existing pitch deck template with the supplied Excel data.
Provide an existing PowerPoint template and source data, then use a prompt such as “Populate this existing pitch deck template with the supplied Excel data.” Before processing, make sure the skill package includes its four referenced files: formatting-standards.md, slide-templates.md, xml-reference.md, and calculation-standards.md. Do not use it to create a presentation from scratch. Review the delivered file in Microsoft PowerPoint because LibreOffice rendering may differ.
What are this skill's strengths and limitations?
- Provides an end-to-end workflow from data extraction and template mapping through validation.
- Requires tabular content to be created as actual PowerPoint table objects rather than tab- or pipe-separated text.
- Includes checks for font consistency, text contrast, slide boundaries, and cross-slide data consistency.
- Addresses common issues involving instruction placeholders, chart sizing, and missing logos.
- Requires an existing PowerPoint template and does not create presentations from scratch.
- Requires user-provided source data and templates; conflicts in source data may still require human judgment.
- LibreOffice may render some PowerPoint fonts, gradients, shape positions, and text wrapping inaccurately.
- The supplied material does not show a skill-specific test suite or an independent platform support matrix.
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 |
|---|---|---|---|---|
| Investment Banking Pitch Deck Filler this page ✓ Anthropic · Official | 53 · Use with care | ★ 40k | 1mo ago | Apache-2.0 |
| Investment Banking Deck Checker ✓ Anthropic · Official | 47 · Use with care | ★ 40k | 1mo ago | Apache-2.0 |
| Investment Banking Strip Profile Builder ✓ Anthropic · Official | 47 · Use with care | ★ 40k | 1mo ago | Apache-2.0 |
| Financial Data Pack Builder ✓ Anthropic · Official | 45 · Not recommended | ★ 40k | 1mo ago | Apache-2.0 |
| XLSX Spreadsheet Skill (Open Cowork) | 69 · Recommended | ★ 2.2k | 12d ago | MIT |
How did FollowSkills review this skill?
The evidence requires backing up the original template, validating source figures, recording footnotes, and warning users to review the delivered file, while limiting the task to populating existing templates. These provide some data-flow transparency and recovery awareness. Points are deducted because user confirmation, sensitive financial-data handling, permission isolation, dependency security, external-source controls, and a complete rollback procedure are not specified; web and other external sources are permitted.
The documentation provides a phased workflow, validation checklists, a three-cycle repair limit, and several failure paths, and the main process is broadly self-consistent. Points are deducted because there are no executable scripts, pinned dependencies, or reproducible tests; LibreOffice/PowerPoint differences, missing templates, and complex XML cases remain dependent on manual judgment, and abnormal-input diagnostics are limited. Static calibration caps this at 10.
The skill clearly targets populating investment-banking PowerPoint templates with data from Excel/CSV and other sources, explicitly excluding presentation creation from scratch. Its decision tree also covers editing and formatting repair. Points are deducted because non-fit boundaries, trigger semantics, input/output contracts, and handling of template or data gaps are not precise enough, and Chinese-language support is undocumented. The core function does not entirely depend on overseas services, so no additional mainland-reachability deduction is applied.
The skill has a clear layered structure covering references, workflow, anti-patterns, error handling, and final checks; the repository context provides an Apache-2.0 license, installation instructions, maintenance/synchronization guidance, and CI for plugin validation and secret scanning. Points are deducted because the skill itself lacks a version, changelog, or clearly assigned maintenance owner, while installation requirements, compatibility details, FAQs, and reference completeness depend on repository context.
The evidence indicates that the skill can map source data into an existing investment-banking template and specifies table objects, formatting consistency, footnotes, boundaries, and delivery warnings, making the intended output potentially usable after review. Points are deducted because no real artifact, test output, or key-path execution evidence is supplied, and visual quality still requires manual PowerPoint review. Static calibration caps this at 7.
The skill includes formulas, checklists, structured validation steps, and an explicit LibreOffice limitation; repository CI adds some auditable plugin and secret checks. Points are deducted because there is no committed test suite covering the skill's key paths, no real execution record, and no independent third-party corroboration. The calculation examples are documentation examples rather than execution evidence. Static calibration caps this at 5.
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 →