improve(research-synthesizer): enhance with phased workflow, tool expansion, and quality verification

[davila7]

Automated Component Improvement

Changes

• Expanded tool list: Added WebSearch and WebFetch with explicit guidance to use them sparingly (only for verifying ambiguous citations or contested claims from upstream outputs)
• Input Discovery Protocol: New section with numbered rules for locating all researcher output files, recording missing researchers in synthesis_metadata.missing_researchers, and never blocking synthesis due to a single absent source
• Phased execution framework: Replaced flat bullet lists with a structured 4-phase workflow — Phase 1 Input Discovery, Phase 2 Parallel Extraction, Phase 3 Cross-Source Integration, Phase 4 Output and Self-Review
• Quality Verification Checklist: Pre-output checklist enforcing citation completeness, contradiction resolution presence, non-empty knowledge_gaps when coverage is incomplete, and single_source labeling
• Dual output requirement: Agent now writes synthesis-summary.md first as a standalone markdown executive summary, then synthesis.json with the full structured data
• Schema additions: Added missing_researchers to synthesis_metadata and single_source to consensus_level enum
• Concrete examples: Replaced abstract domain examples (climate change, AI safety) with realistic Claude Code workflow scenarios referencing actual agent output files (LLM fine-tuning costs, WASM adoption)

Research Summary

The original component had solid structure (JSON schema, contradiction handling, consensus_level) but lacked operational guidance for discovering inputs, handling missing sources, and verifying output quality before writing. The improvements add a complete pre-synthesis protocol and a self-verification gate that make the agent more robust in automated multi-agent pipelines.

Validation

• component-reviewer: PASSED
  • Valid YAML frontmatter with all required fields
  • Kebab-case naming
  • No hardcoded secrets
  • No absolute paths
  • 2 concrete workflow examples with commentary
  • Correct category placement (deep-research-team)

---

Automated review cycle by Component Improvement Loop

---

Summary by cubic

Improves the research-synthesizer with a phased workflow, input discovery, and quality checks to make syntheses more reliable. Also adds limited web verification tools and tightens outputs and schema.

• Area: components (cli-tool/components/). No new components; no docs/components.json regen; no new env vars or secrets.
• Adds a 4-phase workflow (discovery, extraction, integration, output) and an Input Discovery protocol that finds researcher files and continues when some are missing.
• Introduces a Quality Verification checklist (citation completeness, contradiction resolutions, non-empty knowledge_gaps, missing_researchers tracking).
• Expands tools with WebSearch and WebFetch, used only to verify ambiguous or contested claims.
• Requires dual outputs (synthesis-summary.md before synthesis.json) and extends the schema with synthesis_metadata.missing_researchers and consensus_level: single_source.

^{Written for commit 9da5698. Summary will update on new commits.}

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
aitmpl-dashboard	Ready	Preview, Comment	May 1, 2026 8:17pm
claude-code-templates	Ready	Preview, Comment	May 1, 2026 8:17pm

[github-actions]

👋 Thanks for contributing, @davila7!

This PR touches cli-tool/components/** and has been marked review-pending.

What happens next

1. 🤖 Automated security audit runs and posts results on this PR.
2. 👀 Maintainer review — a human reviewer validates the component with the component-reviewer agent (format, naming, security, clarity).
3. ✅ Merge — once approved, your PR is merged to main.
4. 📦 Catalog regeneration — the component catalog is rebuilt automatically.
5. 🚀 Live on aitmpl.com — your component appears on the website after deploy.

While you wait

• Check the Security Audit comment below for any issues to fix.
• Make sure your component follows the contribution guide.

This is an automated message. No action is required from you right now — a maintainer will review soon.

[github-actions]

⚠️ Security Audit Report

Status: ❌ FAILED

Metric	Count
Total Components	763
✅ Passed	360
❌ Failed	403
⚠️ Warnings	1006

❌ Failed Components (Top 5)

Component	Errors	Warnings	Score
vercel-edge-function	3	4	81/100
prompt-engineer	2	0	90/100
neon-expert	2	2	88/100
agent-overview	2	1	89/100
unused-code-cleaner	2	1	89/100

...and 398 more failed component(s)

---

📊 View Full Report for detailed error messages and all components

[cubic-dev-ai]

No issues found across 1 file