deck-review
Scores and strengthens startup pitch decks (pre-seed through Series A) against 35 investor-grade criteria grounded in Sequoia, DocSend, YC, a16z, and Carta data.
npx skills add majiayu000/claude-skill-registry --skill deck-review --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
The deck-review skill scores a startup pitch deck against 35 investor-grade criteria and provides concrete, actionable feedback. It emphasizes a founder-first coaching tone and grounds recommendations in best practices from Sequoia, DocSend, YC, a16z, and Carta data. It produces a structured review with artifacts including a deck inventory, stage profile, slide reviews, and a final report. The process is designed to run inline in the main thread, orchestrating producer scripts and dispatching a deck-review sub-agent for analysis and coaching at specific steps. It requires an attached deck file (PDF, PPTX, markdown, or slide-text) and a user request for review or scoring.
How it works
- Step 1: Read or create founder context to establish company metadata. If missing, it prompts for company name, stage, sector, and geography and initializes founder context.
- Step 2: Ingest the deck into deck_inventory.json by analyzing each slide to capture headline, content summary, visuals description, and word count estimate. It marks input quality issues and handles multi-file or partial decks with appropriate flags. The ingest is performed via a Python script, which validates the output against a JSON schema and injects run_id.
- Step 3: Detect stage via stage_profile.json by evaluating signals (e.g., ARR, funding asks) and AI-company indicators. It records detected_stage, confidence, evidence, and whether the company is AI-focused, along with expected framework and benchmarks. The profile is generated by a Python script and saved with run_id.
- Gate: Confirm Stage and Scope. A gate mechanism may emit a needs_input payload if needed, or read an existing gate_state.json when re-invoked. If needs input, the agent asks the founder via an input prompt and records the answer in gate_state.json, then re-invokes the sub-agent. Outputs include a needs_input payload with gate_state_path, question, options, and context_summary.
- Throughout, the main thread passes run_id to producer scripts to ensure metadata consistency and artifact integrity. The final report is composed after all artifacts are generated and cross-validated.
When to use it
Use ONLY when the user has attached a pitch deck file (PDF, PPTX, markdown, or pasted slide text describing slide-by-slide content) AND has asked for review, scoring, feedback, or critique of the deck. Do not auto-invoke on general fundraising or pitch questions; use ONLY when there is actual deck content to review.
What it can touch
- deck_inventory.json (via deck_inventory.py)
- stage_profile.json (via stage_profile.py)
- gate_state.json (via gate_state.py)
- report.md and related artifacts (via compose_report.py) Note: All artifacts must share the same run_id; the system enforces this via the producer scripts and the final report composer.
Caveats
- The workflow relies on a strict artifact pipeline with schema validation; if any artifact is missing or lacks run_id, a high-level warning may be emitted during composition.
- As a formatting and orchestration-heavy process, it requires the deck content to be parseable for accurate scoring; image-only PDFs or slides with inaccessible text may reduce confidence and influence coaching outputs.
- Re-invocation behavior preserves prior artifacts by rehydrating RUN_ID and skipping steps when appropriate.
# Deck Review Skill Help startup founders strengthen their pitch decks before sending them to investors. Produce a structured, scored review with specific, actionable recommendations grounded in current best practices from Sequoia, DocSend, YC, a16z, and Carta data. The tone is founder-first: a candid coaching session, not a VC evaluation. ## Skill Metadata - **Author:** lool-ventures - **Version:** managed in `founder-skills/.claude-plugin/plugin.json` - **Compatibility:** Python 3.10+ and `uv` for script execution. - **Exports:** - `checklist.json` → `financial-model-review`, `ic-sim`, `fundraise-readiness` ## Skill Execution Model (READ FIRST) This skill runs **inline in the main thread** (not as a sub-agent). The main thread has full tool access including Bash, and is responsible for orchestrating the full pipeline: running producer scripts, persisting artifacts, and dispatching the deck-review sub-agent at specific moments. **Two dispatch contexts for the sub-agent:** - **Context A — Per-step analytical dispatch (Mitigation 1):** Steps 4 and 5 dispatch the deck-review agent via the `Task` tool. The agent does deep analysis and returns structured JSON. The main thread captures
- Skill Metadata
- Skill Execution Model (READ FIRST)
- Input Formats
- Available Scripts
- Available References
- Artifact Pipeline
- Workflow
- Step 0: Path Setup
- Step 1: Read or Create Founder Context
- Step 2: Ingest Deck -> deckinventory.json
- Step 3: Detect Stage -> stageprofile.json
- Gate: Confirm Stage and Scope
- Sub-agent JSON staging (v0.4.2)
- Step 4: Review Each Slide -> slidereviews.json (Context A dispatch)
mkdir -p "$ARTIFACTS_ROOT" Preliminary RUN_ID — used by Step 1 (founder_context init) before slug-aware setup_run.py runs. Will be reused by setup_run via --run-id, OR overwritten by gate_state.json on re-invocation (see below). Resolve REVIEW_DIR (from prompt if provided, else derive) if [ -z "$REVIEW_DIR" ]; then fi mkdir -p "$REVIEW_DIR" mkdir -p "$REVIEW_DIR/.staging" # for ad-hoc sub-agent JSON staging (v0.4.2) if [ -f "$REVIEW_DIR/gate_state.json" ]; then
What does the deck-review skill do?
Scores and strengthens startup pitch decks (pre-seed through Series A) against 35 investor-grade criteria grounded in Sequoia, DocSend, YC, a16z, and Carta data.
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill deck-review --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.