agentleFS
elvismuslielvismusli/business-validator/skills/report-generation/SKILL.md

report-generation

Skill1 starsChanged 9 months agoMITView on GitHub

Use as the final step after all research skills to assemble sections into a complete business validation report with Go/No-Go recommendation and PDF export — requires run_id and all prior sections

  • Installs packages

3 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 5 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Do NOT generate the report until ALL sections exist for the given run_id:

    line 13
  2. Distinctive:

    If any section is missing, inform the user and invoke the missing skill with the run_id.

    line 19
  3. Very specific:

    Check that all 4 section files exist in docs/reports/<run_id>/ using the exact run_id.

    line 60
  4. Distinctive:

    Read each one plus the business brief.

    line 61
  5. Very specific:

    Use the template at @report-template.md as the structure.

    line 65
  6. Distinctive:

    If GO or CONDITIONAL GO, provide 5-7 concrete next steps ordered by priority.

    line 99
  7. Very specific:

    Ensure the output directory exists: mkdir -p docs/reports/<run_id>

    line 103
  8. Very specific:

    If pandoc is not installed, inform the user: "PDF generation requires pandoc.

    line 115
  9. Distinctive:

    If pandoc is installed but the default PDF engine is not available, try:

    line 118

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Report Generation

Overview

Assemble all research sections into a unified business validation report. Add Executive Summary, Go/No-Go recommendation, and generate PDF.

<HARD-GATE> Do NOT generate the report until ALL sections exist for the given run_id: - `docs/reports/<run_id>/01-market-research.md` - `docs/reports/<run_id>/02-competitor-analysis.md` - `docs/reports/<run_id>/03-financial-model.md` - `docs/reports/<run_id>/04-risk-assessment.md`

If any section is missing, inform the user and invoke the missing skill with the run_id. </HARD-GATE>

Inputs

  • run_id: From idea-intake (format: YYYY-MM-DD-<slug>-<hhmm>)
  • Business brief: docs/business-briefs/<run_id>.md
  • All 4 section files in docs/reports/<run_id>/

Output

  • Markdown: docs/reports/<run_id>/REPORT.md
  • PDF: docs/reports/<run_id>/REPORT.pdf

Process

digraph report_generation {
    "Verify all sections exist for run_id" [shape=box];
    "Read all sections" [shape=box];
    "Assemble report from template" [shape=box];
    "Write Executive Summary" [shape=box];
    "Write Go/No-Go recommendation" [shape=box];
    "Write Next Steps" [shape=box];
    "Consolidate sources" [shape=box];
    "Save REPORT.md" [shape=box];
    "Generate PDF" [shape=box, style=filled, fillcolor="#e8f5e9"];

    "Verify all sections exist for run_id" -> "Read all sections";
    "Read all sections" -> "Assemble report from template";
    "Assemble report from template" -> "Write Executive Summary";
    "Write Executive Summary" -> "Write Go/No-Go recommendation";
    "Write Go/No-Go recommendation" -> "Write Next Steps";
    "Write Next Steps" -> "Consolidate sources";
    "Consolidate sources" -> "Save REPORT.md";
    "Save REPORT.md" -> "Generate PDF";
}

Step 1: Verify and Read Sections

Check that all 4 section files exist in docs/reports/<run_id>/ using the exact run_id. Read each one plus the business brief.

Step 2: Assemble the Report

Use the template at @report-template.md as the structure. Replace placeholders with actual section content.

Step 3: Write Original Sections

These sections are NOT research — they are synthesis:

Executive Summary

  • 1 paragraph: What the idea is
  • 1 paragraph: Market opportunity (key numbers)
  • 1 paragraph: Competitive position
  • 1 paragraph: Financial viability
  • 1 paragraph: Key risks and recommendation

Go / No-Go Recommendation

Score the idea on these dimensions:

DimensionScore (1-5)Weight
Market Size and GrowthX25%
Competitive PositionX20%
Financial ViabilityX25%
Risk ProfileX15%
Founder ReadinessX15%
Weighted TotalX.X / 5.0

Decision framework:

  • 4.0+: GO — Strong opportunity
  • 3.0-3.9: CONDITIONAL GO — Proceed with caution, address key risks
  • 2.0-2.9: NO-GO — Significant issues must be resolved first
  • Below 2.0: NO-GO — Fundamental viability concerns

Next Steps

If GO or CONDITIONAL GO, provide 5-7 concrete next steps ordered by priority.

Step 4: Save Report

Ensure the output directory exists: mkdir -p docs/reports/<run_id>

Write to docs/reports/<run_id>/REPORT.md

Step 5: Generate PDF

Attempt PDF generation via pandoc:

pandoc docs/reports/<run_id>/REPORT.md -o docs/reports/<run_id>/REPORT.pdf -V margin-top=25mm -V margin-bottom=25mm -V margin-left=20mm -V margin-right=20mm

If pandoc is not installed, inform the user: "PDF generation requires pandoc. Install with: brew install pandoc (macOS) or apt install pandoc (Linux). The Markdown report is ready at docs/reports/<run_id>/REPORT.md."

If pandoc is installed but the default PDF engine is not available, try:

pandoc docs/reports/<run_id>/REPORT.md -o docs/reports/<run_id>/REPORT.pdf --pdf-engine=weasyprint

Or suggest the user install a LaTeX distribution: brew install basictex

Step 6: Present Results

Show the user:

  1. The run_id
  2. Path to the Markdown report
  3. Path to the PDF (if generated)
  4. The Executive Summary
  5. The Go/No-Go verdict and score

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.