PDF Marketing Report Generator
Compiles website marketing audit data into a client-ready PDF with score gauges, charts, and a prioritized action plan.
Skill only generates a local PDF; data flow is transparent (writes to /tmp then cleans up), sole dependency reportlab is pre-checked and installed, no sensitive-data exfiltration or overreach observed. Deducted because the script source was not provided for review, the curl|bash install pattern is not addressed within the skill, and there is no explicit user-confirmation step.
Docs are detailed and self-consistent on JSON schema, prerequisites, troubleshooting table, demo mode and output verification. Deducted because generate_pdf_report.py source is absent, so key-path behavior, abnormal-input failure feedback and tests cannot be statically verified; capped at 10 by static calibration.
Triggers are precise (slash command and semantic phrases), a PDF-vs-Markdown boundary table and a no-prior-data fallback are provided. Deducted for missing non-fit declarations, English-only output with no Chinese support, and pip/reportlab reachability concerns in constrained networks.
Well-layered progressive disclosure and clear MIT license. Deducted for no versioning or changelog, maintenance responsibility resting only on an unverified publisher, and promotional content in the README.
Claims a directly client-deliverable branded PDF with field-level assembly guidance and concrete scoring methodology, plausibly beating hand-written Markdown. Deducted because output quality is unverified without execution; size/page claims are author assertions, capped at 7.
Scoring methodology, color mapping and JSON schema are auditable primary specifications. Deducted for absence of tests, CI, or third-party execution evidence; sample output is claimed only, so capped at 5.
- generate_pdf_report.py source was not within the review scope; actual PDF output quality and failure behavior are unverified both statically and by execution
- The installer uses curl | bash; prefer cloning and running locally to reduce supply-chain risk
- Output and docs are English-only with no Chinese support; pip-installing reportlab may fail in restricted network environments
- The README contains an external promotional community link unrelated to skill function; evaluate critically
- Model-generated claims such as 'estimated revenue impact' should be manually fact-checked before client delivery
What does this skill do, and when should you use it?
One of 15 skills in the AI Marketing Suite for Claude Code, located at skills/market-report-pdf/. It reads Markdown outputs from the suite's other audit skills (MARKETING-AUDIT.md, SEO-AUDIT.md, etc.), structures them into a fixed JSON schema, and invokes the Python script generate_pdf_report.py to produce a branded PDF. The report includes a circular score gauge, color-coded bar charts for six scoring categories, severity-ranked findings, a three-tier action plan, and an optional competitor comparison table. Best suited to marketing consultants and agencies that need polished client deliverables.
1) Collects existing audit outputs from the project directory (MARKETING-AUDIT.md, LANDING-CRO.md, SEO-AUDIT.md, BRAND-VOICE.md, COMPETITOR-ANALYSIS.md, FUNNEL-ANALYSIS.md, SOCIAL-AUDIT.md, EMAIL-AUDIT.md, AD-AUDIT.md); if none exist, it recommends running /market audit <url> first, or gathers data directly via scripts/analyze_page.py. 2) Assembles a structured JSON payload: URL, brand name, overall score computed from weighted categories (Content 25%, Conversion 20%, SEO 20%, Competitive Positioning 15%, Brand & Trust 10%, Growth & Strategy 10%), an executive summary, 5-10 findings ranked by severity (Critical/High/Medium/Low), a three-tier action plan (this week, 1-3 months, 3-6 months), and up to 3 competitors. 3) Writes the JSON to a temp file, verifies reportlab is installed, runs python3 scripts/generate_pdf_report.py to produce the PDF, verifies the output, and cleans up. Output is MARKETING-REPORT-<domain>.pdf in the project root, typically 5-7 pages and 200-500KB. Competitor data is optional and its section is skipped when absent.
- An agency consultant packaging a full audit of a prospect's website into a printable, emailable PDF before a sales call.
- A freelance marketer finishing a paid audit who needs a polished final report with charts and score visualizations instead of raw Markdown.
- A Claude Code user who has already run /market audit, /market seo, and similar commands and wants the scattered Markdown results consolidated into one report.
- A sales scenario where the report doubles as a prospecting tool: specific findings and an achievable action plan make a compelling free diagnostic.
- Anyone wanting to verify the output quickly after install — running the script with no arguments generates a sample report from placeholder data.
What are this skill's strengths and limitations?
- Genuinely client-facing output: cover page with circular score gauge and A+ to F grade, color-coded bar charts, severity-labeled findings table, numbered action plan, all with consistent script-controlled branding.
- The JSON schema is documented field by field, including scoring guidance for all six categories and the exact weighting formula, so scores are defensible when clients ask why.
- Clear PDF-vs-Markdown decision guidance (PDF for clients, Markdown for internal iteration), and graceful degradation — missing competitor data simply omits that section.
- Demo mode (run with no arguments) produces a sample report, so you can validate the install immediately.
- Requires Python and reportlab; the built-in troubleshooting table itself lists missing modules and JSON parsing errors as known failure modes.
- Report quality is entirely dependent on input data — the skill itself warns 'garbage in, garbage out', and skipping audits noticeably weakens the output.
- Fonts are locked to reportlab's built-in Helvetica and the color palette is preset; no custom branding of fonts or colors is supported.
- No test suite, release history, or user feedback is cited in the source material; the six-category weights are hardcoded, which limits users who want a custom scoring framework.
How do you install this skill?
This skill ships as part of the ai-marketing-claude collection. One-command install:
bash
curl -fsSL https://raw.githubusercontent.com/zubair-trabzada/ai-marketing-claude/main/install.sh | bash
Or manually:
bash
git clone https://github.com/zubair-trabzada/ai-marketing-claude.git
cd ai-marketing-claude
./install.sh
PDF generation needs an extra Python dependency:
bash
pip install reportlab
The repo does not document installing this single sub-skill in isolation.
How do you use this skill?
Recommended workflow: build up data first with /market audit <url>, /market competitors <url>, /market seo <url>, and /market landing <url>, then run /market report-pdf <url> — the skill automatically locates and merges those output files into the report JSON. You can also trigger it by asking for a 'client-ready PDF report'. With no prior data, it either recommends an audit first or analyzes the URL on the spot with scripts/analyze_page.py <url>. The result is MARKETING-REPORT-<domain>.pdf in the project root.