pitch-deck-reviewer
Skill378 starsChanged 56 days agoMITView on GitHub
Reviews pitch decks and investor presentations. Reads slide content, evaluates narrative flow, problem/solution clarity, market sizing, competitive positioning, financial projections, team credibility, and ask clarity. Generates a scored pitch-review.md with slide-by-slide feedback, overall score, top improvements, investor objection predictions, and comparisons to successful decks. Use when reviewing fundraising materials, investor decks, or pitch presentations.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- benyoskovitz/claude-skillstheir own version, first seen Oct 2, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- benyoskovitz/claude-skills★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository229 other files it gives its agents.
- accessibility-auditorSkill
- agent-armySkill
- agent-eval-harnessSkill
- agent-swarm-deployerSkill
- agent-team-builderSkill
- agent-to-agentSkill
- ai-coding-guardrailsSkill
- ai-readiness-assessmentSkill
- animateSkill
- api-documentation-writerSkill
- api-endpoint-scaffolderSkill
- api-load-testerSkill
- athlete-social-media-managerSkill
- before-you-buildSkill
- bi-measure-builderSkill
- board-deck-generatorSkill
- bookkeeping-closeSkill
- bracket-predictorSkill
- brand-consistency-checkerSkill
- brand-voice-analyzerSkill
- browser-test-loopSkill
- budget-optimizerSkill
- cash-flow-forecasterSkill
- champion-identifierSkill
- churn-autopsySkill
- claude-design-criticSkill
- claude-design-system-architectSkill
- claude-landing-composerSkill
- claude-md-writerSkill
- client-health-dashboardSkill
- client-proposal-generatorSkill
- code-review-proSkill
- cold-email-sequence-generatorSkill
- color-palette-extractorSkill
- commercial-lease-analyzerSkill
- company-announcement-writerSkill
- competitor-content-analyzerSkill
- competitor-intel-agentSkill
- competitor-price-trackerSkill
- compliance-checkerSkill
- contact-hunterSkill
- content-repurposerSkill
- context-budgetSkill
- contract-analyzerSkill
- contract-redlinerSkill
- conversation-archaeologistSkill
- cowork-calendar-defragSkill
- cowork-contract-renewal-radarSkill
- cowork-data-room-builderSkill
- cowork-deal-roomSkill
- cowork-expense-auditSkill
- cowork-folder-organizerSkill
- cowork-hiring-screenerSkill
- cowork-home-inventorySkill
- cowork-inbox-triageSkill
- cowork-invoice-chaserSkill
- cowork-medical-bill-auditorSkill
- cowork-qbr-builderSkill
- cowork-rfp-responseSkill
- cowork-sop-writerSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 23
Write slide-by-slide feedback.
- Distinctive: line 31
Verify any market-size, revenue, or growth claim with WebSearch before accepting it.
- Distinctive: line 42
Use investor framing ("This raises a question about...") over verdicts ("This is bad").
- Distinctive: line 44
Write for the founder, to help them improve.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Pitch Deck Reviewer
Review a pitch deck with VC-grade rigor and produce a scored, actionable pitch-review.md.
Contents
references/evaluation-framework.md-- The 8 scored dimensions, weights, red flags, scoring guides, design notes, and overall-score calculation.references/objections-and-benchmarks.md-- Investor objection categories and the canonical successful-deck benchmarks.references/output-template.md-- The exactpitch-review.mdstructure to generate.
Workflow
- Ingest the deck. Determine file type and location, then read every slide. Supported inputs: PDF (Read directly; for 10+ pages read in batches via the
pagesparameter), PowerPoint .pptx (invoke the pptx skill or Read), Google Slides links (WebFetch if public), images PNG/JPG (Read to inspect visually), markdown/text descriptions, and directories of slide images (Glob to enumerate, then Read each). For .pen files, state that they cannot be read directly. - Build a slide inventory. Catalog each slide by number and primary content before any analysis. Identify company name, stage, and industry. Flag any deck over 20 slides as a concern itself; most successful decks run 10-15.
- Score the 8 dimensions. Apply
references/evaluation-framework.md. Score each 1-10 with specific evidence from the deck, then compute the weighted overall score and assign the grade label. - Write slide-by-slide feedback. Cover every slide with content summary, purpose, effectiveness, strengths, weaknesses, and recommendations.
- Predict investor objections. Generate at least 5 using the categories in
references/objections-and-benchmarks.md, each with severity, trigger, suggested verbal response, and deck fix. - Compare to benchmark decks. Reference at least 2 relevant decks from
references/objections-and-benchmarks.md; name where this deck is stronger, weaker, and what to borrow. - Rank the top 5 improvements by impact and list any missing slides or sections for this stage and industry.
- Assemble
pitch-review.mdperreferences/output-template.md. Write the executive summary last. Save it beside the source deck (or the working directory). Show the overall-score math.
Calibration
- Verify any market-size, revenue, or growth claim with WebSearch before accepting it. Never fabricate metrics, comparisons, or statistics.
- Reserve 9-10 on a dimension for work that sets a new standard. Give an overall below 3.0 only when the deck has almost no usable content.
- Adjust for stage: weight pre-product decks toward problem/solution clarity, team, and market; relax traction and financials expectations and note the calibration.
- Score absent major sections (e.g., missing financials or team) 1-2 and flag them in the executive summary and the missing-sections list. Review works in progress rather than refusing.
- Evaluate intentionally unconventional decks (memo-style, demo-first, single-page) on their own terms while noting the divergence from convention.
- For multiple versions, review the latest and note which changes improved the deck and which regressed.
Voice
- Be direct and specific. Replace vague praise with concrete observations.
- Pair every critique with a concrete fix.
- Use investor framing ("This raises a question about...") over verdicts ("This is bad").
- Acknowledge what works before identifying gaps. Define any jargon used.
- Write for the founder, to help them improve.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

