config: add k-horikawa-dale/auto-report-creator to Agent Skills & AGENTS.md collection

[k-horikawa-dale]

Adds one repository to the existing 10124.agent-skills.yml collection.

Repository

k-horikawa-dale/auto-report-creator — turns chat transcripts (ChatGPT/Claude exports, JSON, JSONL, or labeled text) into structured reports: executive briefs, technical write-ups, meeting minutes, changelogs, FAQs, or short digests, as Markdown, HTML or JSON.

Why it fits this collection

It ships as an agent skill (SKILL.md plus bundled scripts and references) designed to be dropped into an agent's skills directory, which is what this collection tracks. It also works standalone as a CLI.

Two properties that distinguish it from general summarization tools:

• Extractive by design. Every sentence in a generated report is copied verbatim from the source transcript and carries a ^[message N] citation. Nothing is paraphrased or invented, and a test enforces this rather than leaving it as a documented promise.
• Secret redaction runs before analysis, so API keys pasted mid-debug cannot reach a shared report. False positives are handled (Luhn check for card numbers, allowlists for localhost and example.com).

Zero third-party dependencies, standard library only, Python 3.9+. English and Japanese output with automatic language detection.

Verification

• Repository is public and resolves with HTTP 200 from the GitHub API
• Name matches the required owner/repo format, no duplicate entry in the collection
• Collection file parses with the same logic as scripts/verify-collection.mjs (id and name unchanged, 13 unique items)
• CI on the added repository is green across Linux, macOS and Windows on Python 3.9–3.13

Happy to move this to a different collection or close the PR if it is not a fit.

[vercel]

@k-horikawa-dale is attempting to deploy a commit to the pingcap Team on Vercel.

A member of the Team first needs to authorize it.