Skip to content

[TypeSpec Authoring][PoC] Agentic search only approach #16267

Draft
haolingdong-msft wants to merge 9 commits into
mainfrom
typespec-authoring-agentic-search-only-clean
Draft

[TypeSpec Authoring][PoC] Agentic search only approach #16267
haolingdong-msft wants to merge 9 commits into
mainfrom
typespec-authoring-agentic-search-only-clean

Conversation

@haolingdong-msft

Copy link
Copy Markdown
Member

Fix #15676

haolingdong-msft and others added 7 commits June 18, 2026 17:38
Squashed net changes of typespec-authoring-agentic-search-only vs main:
- azure-typespec-author SKILL.md + references (agentic search only; drop authoring-plan MCP tool)
- evaluate/ eval cases and consolidated forced/no-skill/trigger suites
- fixtures: Widget package.json/package-lock.json; gitignore result/

Co-authored-by: JoyerJin <joyerjin@microsoft.com>

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
…entic-search-only-clean

# Conflicts:
#	.github/skills/azure-typespec-author/evaluate/evals/002002.eval.yaml
#	.github/skills/azure-typespec-author/evaluate/evals/003001.eval.yaml
#	.github/skills/azure-typespec-author/evaluate/evals/003002.eval.yaml
#	.github/skills/azure-typespec-author/evaluate/evals/005001.eval.yaml
#	.github/skills/azure-typespec-author/evaluate/suites/forced.eval.yaml
#	.github/skills/azure-typespec-author/evaluate/suites/no-skill.eval.yaml
#	.github/skills/azure-typespec-author/references/validation.md
…RL, add 002011 eval coverage

- Remove ARM versioning URL (https://azure.github.io/typespec-azure/docs/howtos/arm/versioning/)
  which now redirects to the already-listed evolving-apis page
- Update doc-coverage.md with new eval 002011 (checkExistence HEAD op)
  mapped to Case 2 — Add Resource Operations (ARM)
- All 8 case headings present and non-empty
- All URLs verified HTTP 200
- 0 eval cases with missing related docs

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Add 4 new authoritative URLs across existing cases:
- Case 2: ARM step 04 (custom actions walkthrough)
- Case 3: Uncommon scenario for converting specs with multiple previews
- Case 6: TypeSpec core scalars reference
- Case 7: Azure Core decorators reference

Fix Case 2 step 05 link text (was mislabeled as 'Custom Actions',
now correctly says 'long-running resource operations').

All 51 URLs verified to return HTTP 200.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
@github-actions

github-actions Bot commented Jul 6, 2026

Copy link
Copy Markdown

📊 GEPA Skill Quality Scores

Skill Quality Triggers Tests
azure-typespec-author ⚠️ 0.57 N/A ---
sensei ⚠️ 0.57 N/A ---
azsdk-common-prepare-release-plan ✅ 0.86 N/A ---
azsdk-common-apiview-feedback-resolution ✅ 1.00 N/A ---
azsdk-common-generate-sdk-locally ✅ 1.00 N/A ---
azsdk-common-pipeline-troubleshooting ✅ 1.00 N/A ---
azsdk-common-sdk-release ✅ 1.00 N/A ---
markdown-token-optimizer ✅ 1.00 N/A ---
sdk-ai-bot-eval-dataset ✅ 1.00 N/A ---
sdk-ai-bot-run-evaluation ✅ 1.00 N/A ---
skill-authoring ✅ 1.00 N/A ---

9/11 skills at quality ≥ 0.80

How to improve
# Score a specific skill
python .github/skills/sensei/scripts/gepa/auto_evaluator.py score --skill <name> --skills-dir .github/skills --tests-dir tests

# Optimize a skill with GEPA
python .github/skills/sensei/scripts/gepa/auto_evaluator.py optimize --skill <name> --skills-dir .github/skills --tests-dir tests

…entic-search-only-clean

# Conflicts:
#	.github/skills/azure-typespec-author/evaluate/suites/forced.eval.yaml
#	.github/skills/azure-typespec-author/evaluate/suites/trigger.eval.yaml
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.

[TypeSpec Authoring] PoC: agent summarize case-based references automatically and decide between agentic search vs. KB

1 participant