Feng GAO
Community@gaofeng21cn · Guangzhou, China
Professor of Medical AI @ SYSU. Building clinically grounded AI, agent-first research systems, and auditable workflow infrastructure.
Agent Skills by Feng GAO
Showing 27 vetted skills indexed across 2 GitHub repositories.
gaofeng21cnmas-scholar-skills
Sync medical research skill modules from the canonical mas-scholar-skills repository to OPL workspaces.
gaofeng21cnopl-domain-progress-transition-reviewer
Review OPL domain progress transition packets for evidence gaps and routing accuracy.
gaofeng21cnopl-completion-audit-writer
Audit OPL plan items against evidence to map completion statuses and flag blockers.
gaofeng21cnopl-agent-package-trust-reviewer
Review OPL agent package trust evidence and classify risk levels.
gaofeng21cnopl-charter-authority-reviewer
Classify statement authority types and flag forbidden claims in OPL foundation materials.
gaofeng21cnopl-memory-artifact-lifecycle-curator
Compile existing memory and artifact refs into structured lifecycle briefs.
gaofeng21cnopl-atlas-capability-router
Resolve ambiguous OPL Atlas catalog references and generate refs-only route packets.
gaofeng21cnopl-pack-admission-reviewer
Review OPL pack admission candidates for registry fit and contract compliance.
gaofeng21cnopl-stagecraft-stage-designer
Design OPL Stagecraft stage configurations with explicit authority boundaries and handoff evidence requirements.
gaofeng21cnopl-foundry-promotion-reviewer
Review OPL Foundry Lab promotion candidates against scorecards, work orders, and rollback evidence.
gaofeng21cnopl-external-runtime-provider-fit-reviewer
Review external runtime provider fit for OPL executor and stage substrate requirements.
gaofeng21cnopl-source-module-boundary-reviewer
Detect OPL source module boundary violations and classify them for code review.
gaofeng21cnopl-connect-connector-receipt-auditor
Audit OPL Connect connector receipt candidates for evidence completeness and gap classification.
gaofeng21cnopl-ledger-evidence-curator
Assess evidence sufficiency and classify gaps for OPL work claims.
gaofeng21cnopl-runway-compute-operator
Diagnose OPL Runway compute setup issues and classify execution failures across compute paths.
gaofeng21cnopl-foundry-agent-improver
Analyze OPL Foundry Lab agent failures and plan minimal rewrites.
gaofeng21cnopl-workspace-source-readiness-auditor
Audit OPL Workspace source material and classify consumption-blocking gaps.
gaofeng21cnopl-workspace-handoff-writer
Generate five OPL handoff packet types with strict truth adherence.
gaofeng21cnopl-stage-quality-gate-critic
Identify rubric gaps, evidence insufficiencies, evaluator overclaims, and owner route flaws in OPL stage quality gates.
gaofeng21cnopl-eval-harness-designer
Design structured evaluation harnesses for OPL Foundry Lab candidates.
gaofeng21cnopl-pack-capability-reviewer
Review OPL declarative packs for boundary defects and ABI mismatches.
gaofeng21cnopl-runway-recovery-playbook-writer
Generate structured recovery playbooks for OPL Runway failures.
gaofeng21cnopl-incident-root-cause-triager
Diagnose OPL operational incidents and generate owner-routable root-cause briefs.
gaofeng21cnopl-connect-source-and-skill-router
Route OPL Connect source and skill tasks while enforcing domain authority boundaries.