Skills 2026-05-11 13:57 UTC
claude-superskills
Committed by 匿名贡献者. Update Category: New Features & Capability Expansion. Commit details: feat(deep-research): add iterative gap analysis, research plan phase, and named critique passes. This update introduces new capabilities and expands supported use cases.
CommitRaw commit message
feat(deep-research): add iterative gap analysis, research plan phase, and named critique passes - Enhancement A: Research Plan Phase (Step 0.5) — create and show a structured research plan with dimensions, source strategy, and known unknowns before launching agents - Enhancement B+C: Gap Analysis & Iteration Check — after each agent batch, reason about coverage vs gaps, decide whether to run targeted follow-up iterations; includes explicit stop criteria (5 conditions) - Enhancement D: Named Self-Critique Passes — 3 named passes before synthesis: Factual Accuracy, Completeness, Coherence & Bias - Enhancement E: Limitations & Caveats section in report-template.md — dedicated section separate from Confidence & Gaps Patterns adopted from frank-luongt/faos-skills-marketplace deep-research skill; existing architecture (agent topology, Evidence Ledger, Frontier mode, etc.) kept intact. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>