Skip to content

feat: add clarifying questions to pptx, pdf, xlsx skills#110

Merged
simongdavies merged 2 commits intohyperlight-dev:mainfrom
simongdavies:skill-clarifying-questions
May 7, 2026
Merged

feat: add clarifying questions to pptx, pdf, xlsx skills#110
simongdavies merged 2 commits intohyperlight-dev:mainfrom
simongdavies:skill-clarifying-questions

Conversation

@simongdavies
Copy link
Copy Markdown
Member

Each skill now has a structured 'Clarifying Questions' section that tells the LLM exactly what to ask when details are missing:

  • pptx-expert: topic, audience, slide count, purpose, theme, charts
  • pdf-expert: document type, topic, audience, sections, page format
  • xlsx-expert: purpose, data structure, data source, charts, formulas

Questions are grouped into three tiers:

  1. Always needed (ask if missing)
  2. Ask if relevant to the request
  3. Never ask (use sensible defaults)

The LLM is instructed to group questions into ONE ask_user call, skip anything already specified, and offer defaults the user can accept with 'yes'.

Each skill now has a structured 'Clarifying Questions' section that
tells the LLM exactly what to ask when details are missing:

- pptx-expert: topic, audience, slide count, purpose, theme, charts
- pdf-expert: document type, topic, audience, sections, page format
- xlsx-expert: purpose, data structure, data source, charts, formulas

Questions are grouped into three tiers:
  1. Always needed (ask if missing)
  2. Ask if relevant to the request
  3. Never ask (use sensible defaults)

The LLM is instructed to group questions into ONE ask_user call,
skip anything already specified, and offer defaults the user can
accept with 'yes'.

Signed-off-by: Simon Davies <simongdavies@users.noreply.github.com>
Copilot AI review requested due to automatic review settings May 7, 2026 08:43
@simongdavies simongdavies added the enhancement New feature or request label May 7, 2026
Copy link
Copy Markdown
Contributor

Copilot AI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

Adds a standardized “Clarifying Questions” section to the PPTX/PDF/XLSX skills to drive consistent requirement gathering before file generation, and updates each skill’s setup sequence to explicitly start with clarification via ask_user.

Changes:

  • Introduce a structured “Clarifying Questions” section with Always needed / Ask if relevant / Never ask tiers.
  • Update setup sequences to begin with a “Clarify requirements” step referencing the new section.
  • Provide suggested defaults users can accept in a single consolidated ask_user interaction.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 3 comments.

File Description
skills/xlsx-expert/SKILL.md Adds clarifying-question tiers and updates setup sequence to start with requirements clarification.
skills/pptx-expert/SKILL.md Adds clarifying-question tiers (topic/audience/slide count/purpose/etc.) and updates setup sequence accordingly.
skills/pdf-expert/SKILL.md Adds clarifying-question tiers (doc type/sections/page format/etc.) and updates setup sequence accordingly.

Comment thread skills/xlsx-expert/SKILL.md
Comment thread skills/pptx-expert/SKILL.md Outdated
Comment thread skills/pdf-expert/SKILL.md Outdated
…sset handling

All 3 skills mentioned URLs/images in clarifying questions but the
setup sequence only applies file-builder (fs-write), leaving the LLM
unable to actually fetch remote content.

- xlsx: data source URL now notes to apply web-research profile
- pptx: images changed from URLs to local paths, with web-research
  guidance if remote URLs are needed
- pdf: branding logo changed from URL to local path/bytes, with
  web-research guidance if remote URLs are needed

Signed-off-by: Simon Davies <simongdavies@users.noreply.github.com>
@simongdavies simongdavies merged commit c6ed248 into hyperlight-dev:main May 7, 2026
12 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

enhancement New feature or request

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants