SKILL RADAR · GITHUB
Useful agent skills, ranked by real adoption.
Browse verified skills from public repositories. Compare what each skill does, package contents, install activity, and source evidence before adding it to your agent.
- CATALOG SCOPE
- VERIFIED
- skills
- 5,771
- repositories
- 824
- last synced
- Sep 8, 2026
robsonrungcodex-runner
Execute prompts using Codex CLI in non-interactive exec mode. Use when users explicitly request Codex execution, when a workflow needs a Codex CLI run inside this repository, or when a cross-runner workflow selects Codex as the preferred model and native Codex subagents are unavailable.
- installs
- 2
- GitHub stars
- 2
- Updated
- Sep 8
robsonrungcoding-design-plan
Plan the implementation shape of one approved coding task. Use when a module, data, API, component, or deployment boundary still needs a design decision; use to-tasks for task decomposition.
- installs
- 2
- GitHub stars
- 2
- Updated
- Sep 8
robsonrungcoding-review-simplify
Review a completed code change for behavior-preserving simplification, local correctness, and maintainability. Use after implementing a scoped task or when the user asks to simplify a concrete diff. Do not use as a full feature review or to choose the delivery reviewer plan.
- installs
- 2
- GitHub stars
- 2
- Updated
- Sep 8
robsonrungcollaborative-delivery
Deliver an approved task through an auditable multi-model panel with test, review, and verification evidence. Use when the user requires a recorded panel trail for implementation. For the normal fast path, use implement-and-review or implement-tasks.
- installs
- 2
- GitHub stars
- 2
- Updated
- Sep 8
robsonrungconsensus-summary-html
Render an existing models-consensus result as a self-contained HTML decision brief. Use when asked to visualize a council report; preserve its opinions and evidence without running a new council.
- installs
- 2
- GitHub stars
- 2
- Updated
- Sep 8
robsonrungdata-systems-coding-lens
Review implementation risks in persistent state and asynchronous data paths. Use for schemas, migrations, queues, caches, retries, or concurrent writes; use macro-architecture to decide service and data ownership.
- installs
- 2
- GitHub stars
- 2
- Updated
- Sep 8
robsonrungdcode-runner
Execute prompts using DeepAgents CLI (dcode) non-interactive mode with the user's already-configured model and credentials. Use only when the user explicitly requests dcode or DeepAgents execution. It is not an approved implementation or review route.
- installs
- 2
- GitHub stars
- 2
- Updated
- Sep 8
robsonrungdecide-about-disagreements
Help the user resolve material disagreements in an existing models-consensus result. Present recorded opinions, distinct options, and a recommendation for each open decision without rerunning the council or implementing it.
- installs
- 2
- GitHub stars
- 2
- Updated
- Sep 8
robsonrungdesign-gate
Select relevant design lenses for a planned slice and return a proceed-or-revise verdict. Use during task planning or when a design boundary changes; invoke a named lens directly for a single-lens review.
- installs
- 2
- GitHub stars
- 2
- Updated
- Sep 8
robsonrungdesign-patterns
Choose or review a Gang-of-Four pattern when recurring code structure creates a concrete design problem. Use to compare a named pattern with a simpler design; use architecture lenses for system boundaries.
- installs
- 2
- GitHub stars
- 2
- Updated
- Sep 8
robsonrungdiagnose
Find and prove the cause of a bug, failing test, or unexpected behavior. Use for unexplained failures, including pipeline verification failures; an understood failure within a TDD loop stays with tdd.
- installs
- 2
- GitHub stars
- 2
- Updated
- Sep 8
robsonrungdistributed-systems-patterns
Choose or review distributed topology and event-stream patterns. Use for replicas, shards, container helpers, queues, or event-driven adoption and migration; use macro-architecture for overall system style.
- installs
- 2
- GitHub stars
- 2
- Updated
- Sep 8
robsonrungdiverse-plan
Compare independent implementation approaches and produce one reviewed plan. Use when the user explicitly requests a diverse plan or several model-generated approaches; it does not implement the plan.
- installs
- 2
- GitHub stars
- 2
- Updated
- Sep 8
robsonrungdomain-driven-design
Review domain boundaries, business language, aggregates, and domain events. Use when a change crosses bounded contexts or restructures business rules within one context.
- installs
- 2
- GitHub stars
- 2
- Updated
- Sep 8
robsonrungdynamic-harness
Plan and run a bounded worker fleet for a complex task. Use when the user explicitly requests delegation, parallel work, a tournament, or a managed mission.
- installs
- 2
- GitHub stars
- 2
- Updated
- Sep 8
robsonrungexplain-architecture
Create a self-contained HTML orientation page explaining a codebase or subsystem with a map and one traced flow. Use for onboarding or architecture overviews; use html-explainer for a detailed code walkthrough.
- installs
- 2
- GitHub stars
- 2
- Updated
- Sep 8
robsonrungfable-mindset
Apply the Fable engineering posture to intake, diagnosis, decisions, implementation, and reports. Use when the user requests this mindset or an assessment of that work process; it does not select a model.
- installs
- 2
- GitHub stars
- 2
- Updated
- Sep 8
robsonrungfrontend-design
Build a custom frontend interface from a chosen visual direction. Use for a bespoke page, component, or site; use ui-ux-pro-max for reusable design systems and advanced-react for React behavior.
- installs
- 2
- GitHub stars
- 2
- Updated
- Sep 8
robsonrungfull-review
Review code changes or requirements with evidence-backed findings. Use for integration seams, public contracts, security, migrations, or a requested deep review; keep local task reviews scoped.
- installs
- 2
- GitHub stars
- 2
- Updated
- Sep 8
robsonrunggemini-runner
Execute prompts using Antigravity CLI (agy) headless print mode for a Gemini/Google seat. Use when users request Gemini execution, Antigravity CLI execution, or when a consensus workflow needs a Gemini seat and local agy is installed.
- installs
- 2
- GitHub stars
- 2
- Updated
- Sep 8
robsonrunggrok-runner
Execute prompts using Grok CLI in headless print mode as the xAI seat (Grok 4.6). Use when users explicitly request Grok execution, when a multi-model workflow needs an xAI seat for provider diversity, or when a cross-runner workflow selects Grok as the preferred model.
- installs
- 2
- GitHub stars
- 2
- Updated
- Sep 8
robsonrunghtml-explainer
Create a self-contained HTML walkthrough of a subsystem or flow, with diagrams and expandable source evidence. Use for detailed interactive code explanations; use explain-architecture for a short orientation.
- installs
- 2
- GitHub stars
- 2
- Updated
- Sep 8
robsonrungimplement-and-review
Implement and review one approved coding task through approved model and effort routes. Use for a scoped task with an acceptance contract; use implement-tasks to coordinate a queue.
- installs
- 2
- GitHub stars
- 2
- Updated
- Sep 8
robsonrungimplement-tasks
Execute an approved task queue with model approval, dependency scheduling, and integration review. Use when the user asks to implement approved tasks; implement-and-review handles each task.
- installs
- 2
- GitHub stars
- 2
- Updated
- Sep 8