Skip to content

Add Deep Company Research to the Extruct API skill#5

Merged
persiyanov merged 3 commits into
mainfrom
feat/deep-research-skill
Jun 10, 2026
Merged

Add Deep Company Research to the Extruct API skill#5
persiyanov merged 3 commits into
mainfrom
feat/deep-research-skill

Conversation

@persiyanov

Copy link
Copy Markdown
Contributor

What

Extends the extruct-api skill with the new Deep Company Research API (extruct-ai/extruct-ai#217):

  • CLI (scripts/extruct-api): new deep-research resource — create (payload validation: non-empty brief, depth enum, output_schema object), get, list (paginated), and poll (terminal on done/failed, defaults 15s × 240 attempts for minutes-long runs). Mirrors the deep-search command group exactly (error JSON shape, exit codes, pagination checks).
  • SKILL.md: new "Deep Company Research" section — when to use it vs Deep Search/tables, Pro-gating preflight (deep-research create added to Pro-only actions), depth/credit semantics (25/50/75 budgets, full budget required to start, billed per completed agent, failed tasks refund), and how to read reports: kind, citations, per-field basis, always surface degradation_reasons, relay failure_reason (rejected briefs include fix-it suggestions).
  • references/deep-company-research.md: playbook for brief-writing, depth choice, and output-schema design.
  • openai.yaml short_description updated; marketplace bumped 1.0.6 → 1.0.7.

Validation

  • python3 scripts/validate_repo.py
  • CLI smoke: deep-research --help + all subcommand parsers ✅; payload/pagination validation paths exercised offline ✅
  • No live calls — the endpoints are not deployed yet.

Note

Draft until the feature ships (extruct-ai/extruct-ai#217). Merge after deploy.

🤖 Generated with Claude Code

New deep-research CLI resource (create/get/list/poll, mirroring deep-search:
payload validation, pagination checks, poll loop sized for minutes-long runs
at 15s x 240 attempts), a SKILL.md section covering when to use it vs Deep
Search, Pro gating, depth/credit semantics, and how to read reports
(citations, basis, degradation_reasons, failure_reason), plus a
deep-company-research reference playbook for brief-writing, depth choice,
and output-schema design. Marketplace bumped to 1.0.7.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
@persiyanov persiyanov marked this pull request as ready for review June 10, 2026 18:53
persiyanov and others added 2 commits June 10, 2026 21:01
People are first-class targets ("research this lead and the team they work
with"), so the section, routing, playbook, and examples drop the
company-only framing: renamed references/deep-research.md, added a
people-brief example to SKILL.md and the playbook, a people typical-ask,
and a cross-reference from the Research People section.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
@persiyanov persiyanov merged commit 84080f7 into main Jun 10, 2026
2 checks passed
@persiyanov persiyanov deleted the feat/deep-research-skill branch June 10, 2026 19:20
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant