report-generator
Skill78 starsChanged 23 days agoMITView on GitHub
Assemble report.md, executive_summary.md and report.pdf from the artifacts. Every number is registered against its source; the PDF is the primary deliverable and the Markdown is the citable source.
15 other projects use a version of this.
Adopted by
Most recently taken up by
- tomwangowa/agent-skillstheir own version, first seen Oct 21, 2026
- Sigma-Works/iCode-Claude-Skillstheir own version, first seen Oct 20, 2026
- NVIDIA/SkillSpectortheir own version, first seen Oct 13, 2026
- nek1987/auto-agent-harnesstheir own version, first seen Oct 11, 2026
- tathadn/data-quality-skills-pipelinetheir own version, first seen Oct 8, 2026
- dotlabshq/cruxtheir own version, first seen Oct 8, 2026
Other versions15 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 15
- NVIDIA/SkillSpector★ 25ksame name
- claude-office-skills/skills★ 433same name
- huifer/claude-code-seo★ 110same name
- ellenfengyunjing/ecommerce-product-research-skills★ 19same name
- o2alexanderfedin/strategic-research-automation-template★ 19same name
- HK-hub/AgentSkills★ 6same name
- dotlabshq/crux★ 4same name
- nek1987/auto-agent-harness★ 4same name
- Sigma-Works/iCode-Claude-Skills★ 3same name
- tomwangowa/agent-skills★ 2same name
5 more not listed
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 repository10 other files it gives its agents.
- data-masterSkill
- data-cleanerSkill
- data-explorerSkill
- data-profilerSkill
- data-routerSkill
- data-transformerSkill
- data-validatorSkill
- deep-analyzerSkill
- statistical-analyzerSkill
- visualizerSkill
The file
We split this skill into the rules it gives an agent
6 rules. 0 are common, 6 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 8 edited versions we read keep these.
6
Distinctive
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 17
After figures render. Never before the statistical gates have artifacts to read -- a report written first is a report that has to be edited …
- Distinctive: line 41
every number in the prose appears in the provenance table.
- Distinctive: line 42
every forbidden construction is absent outside a disclaimer.
- Distinctive: line 57
do not type a number into the report that was not read from an artifact.
- Distinctive: line 58
do not treat a BLOCKED PDF as success -- the markdown is complete, the deliverable is not.
- Distinctive: line 60
do not duplicate the executive summary's argument in the full report with different wording.
“Common” is measured: kept by at least 6 in 10 of the 8 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
report-generator
Stage 3b of the data-master pipeline. The router in ../../SKILL.md owns the
end-to-end run; this skill owns one stage and one gate. Read the parent contract before
changing anything here: the invariants in ../../SKILL.md (additive-only, types decided
once, default-FAIL gates, three-part provenance) are not restated, they are inherited.
When to invoke
After figures render. Never before the statistical gates have artifacts to read -- a report written first is a report that has to be edited to fit the numbers.
Inputs
- all of
outputs/analysis/*.json outputs/cleaned/cleaning_run.jsonoutputs/charts_rendered.json
Outputs
report.md-- full report, 6 sections + 4 appendicesexecutive_summary.md-- one page, same numbersreport.pdf-- A4, CJK-capable, page numbers, table of contentsreport_build.json-- provenance for every number, PDF backend trail, self-audit
Commands
python scripts/assemble_report.py --raw <raw.csv> --outdir outputs --title "<title>"Gate (default FAIL)
- every number in the prose appears in the provenance table;
- every forbidden construction is absent outside a disclaimer;
- the PDF cascade records which backend ran and why earlier tiers were skipped;
report.mdis complete even when the PDF is BLOCKED.
Failure modes seen in real runs
- Self-referential hash. A document cannot contain its own sha256; the report
points at
report_build.jsoninstead of printing a stale value. - A PDF backend that "works" but drops images. The browser tier embeds images as
data URIs because
file://sandboxing silently loses them. - A missing CJK font rendered as black boxes. The reportlab tier falls back to a CID font and records whether it was embedded.
Do not
- do not type a number into the report that was not read from an artifact;
- do not treat a BLOCKED PDF as success -- the markdown is complete, the deliverable is not;
- do not duplicate the executive summary's argument in the full report with different wording.
Upstream / downstream
- previous:
../visualizer/SKILL.md - next:
../data-validator/SKILL.md - templates:
../../references/report-template-catalog.md
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.