91 skills · sorted by relevance
Review agent work histories to identify recurring failures, repeated discoveries, and stalled tasks. Improve the prompts, specialist agents, skills, tools, and memory that guide future work, then submit those changes separately for review.
Builds, runs, debugs, and operates event-driven applications using EventBridge Event Bus - a managed, centrally governed publish/subscribe event bus that an organization can share across many teams and accounts. Applicable when workloads need event-driven architectures, decoupling, choreography, asynchronous integration, pub/sub, fan-out, event ordering, delivered in sequence, event router, event broker, event bus, event store, event retention, event replay, deduplication, avro, protobuf, cloudevents, uses events or messages. Designed as a centrally managed event bus platform allowing governance and control - such as subscriber control, revocation, per-account throttling - while giving application owners flexibility across an organization spanning multiple accounts, with end-to-end open-standards observability and cost allocation. Also helps reduce costs in high fan-out scenarios where multi-account forwarding adds cost. A serverless, managed alternative to self-managed event streaming platforms.
The English summary is being prepared. Showing the original description.
Use when the user mentions Higgsfield Canvas, a node-based or node graph workspace, an infinite board/canvas, chaining generations into a pipeline, or wants to wire prompts → images → videos across models on one surface. Covers what Canvas is, the node categories, the seven models that run inside Canvas, the named canvas patterns (Simple Seedance, Extend Video, Image Edit, StoryBoard With Elements, Long Video fan-out), the build-free / generate-paid cost model, reusable templates, assets-as-nodes, and Shared Canvas live collaboration. Also trigger on 'Higgsfield ComfyUI alternative', 'node workflow', or 'connect nodes to build a scene/campaign'.
The English summary is being prepared. Showing the original description.
Decision protocol for the map-reduce / dynamic fan-out pattern in LM pipelines — "given list L, run f(item) for each item in parallel, then combine". Activates when the coder agent is about to process N items with N LM calls (per-doc summarize, per-query retrieve, per-candidate rank, parallel tool fan-out). Encodes the *when*, *how many at once*, *what to do when one fails*, and *how to reduce* — not the API of any single framework. Cross-framework: LangGraph `Send`, CrewAI parallel tasks / Flow, `asyncio.gather`, `ThreadPoolExecutor`, LlamaIndex batch retrieval.
The English summary is being prepared. Showing the original description.
Use when designing event-driven workflows, decoupling services, implementing fan-out patterns (one trigger, many downstream handlers), implementing idempotent event handling with IDs (24-hour dedupe window), or handling at-least-once delivery from external sources like Stripe webhooks. Covers Inngest event schema, payload format, naming conventions, IDs for idempotency, the ts param, fan-out patterns, and system events like inngest/function.failed.
The English summary is being prepared. Showing the original description.
Orchestrate a MULTI-PHASE, dependent, or resumable run over many provider subagents from a JS script, off the main context (`cc-fleet workflow`). Use for fan-out→barrier→synthesis, per-item pipelines, loop-until-dry, or a run that must survive a kill and `--resume` from its journal. NOT a flat fan-out of independent tasks (that is /cc-fleet:subagent — cheaper, no script); NOT interactive collaboration (that is /cc-fleet:team); NOT trivial single-shot work for the main session.
The English summary is being prepared. Showing the original description.
Orchestrate a MULTI-PHASE, dependent, or resumable run over many cc-fleet PROVIDER subagents from a JS script, off the main context (`cc-fleet workflow`). Use for fan-out→barrier→synthesis, per-item pipelines, loop-until-dry, or a run that must survive a kill and `--resume` from its journal. Trigger ONLY to run a cc-fleet provider workflow. NOT a flat fan-out of independent tasks (that is /cc-fleet:subagent — cheaper, no script); NOT trivial single-shot work; NOT for editing or researching cc-fleet's own workflow code.
The English summary is being prepared. Showing the original description.
Shapes for running many sub-agents at once — fan-out with a barrier, maker/checker, generate-and-filter, tournaments for ranking, hypothesis fan-out for root cause, and loops that stop on a condition with a cap. Use when a phase has several independent targets, when a task is too large or too repetitive for one window, when the same edit applies across many files, when ranking or judging a large set, when several hypotheses compete, or when work must repeat until a signal says stop. Does not apply when units depend on each other's results or the cause is still unknown — investigate first, then parallelize.
The English summary is being prepared. Showing the original description.
Parallel agent orchestration for OST exploration. Fan-out multiple solution explorations, fan-in results to compare and select winners.
The English summary is being prepared. Showing the original description.
Run a read-only parallel Kimi review fan-out across many independent targets. Use only for explicit broad fan-out requests where one subagent per target is the point.
The English summary is being prepared. Showing the original description.
Lightweight build: explore → plan → approve → implement in a single Opus context, no sub-agent fan-out. Verifies build + tests, optionally commits, and offers a separate /code-review pass at the end. Use by default for everyday features; reach for /build only when the work needs parallel fan-out or exceeds one context.
The English summary is being prepared. Showing the original description.
Lightweight refactor: audit → plan → approve → implement in a single Opus context, no sub-agent fan-out. Behavior preservation is the goal — verifies tests still pass, optionally commits, and offers a separate /code-review pass at the end. Use by default; reach for /refactor only when the cleanup needs parallel fan-out or exceeds one context.
The English summary is being prepared. Showing the original description.
Lightweight refactor: audit → plan → approve → implement in a single Opus context, no sub-agent fan-out. Behavior preservation is the goal — verifies tests still pass, optionally commits, and offers a separate /code-review pass at the end. Use by default; reach for /claude-setup:refactor only when the cleanup needs parallel fan-out or exceeds one context.
The English summary is being prepared. Showing the original description.
Lightweight build: explore → plan → approve → implement in a single Opus context, no sub-agent fan-out. Verifies build + tests, optionally commits, and offers a separate /code-review pass at the end. Use by default for everyday features; reach for /claude-setup:build only when the work needs parallel fan-out or exceeds one context.
The English summary is being prepared. Showing the original description.
Use when designing event-driven workflows, decoupling services, implementing fan-out patterns (one trigger, many downstream handlers), implementing idempotent event handling with IDs (24-hour dedupe window), or handling at-least-once delivery from external sources like Stripe webhooks. Covers Inngest event schema, payload format, naming conventions, IDs for idempotency, the ts param, fan-out patterns, and system events like inngest/function.failed.
The English summary is being prepared. Showing the original description.
Conversation patterns and interaction protocols for multi-agent systems. Covers request/response, pub/sub, blackboard, delegation chains, debate, critique, consensus, fan-out/fan-in, supervisor-worker, and peer negotiation. Deep analysis of AutoGen conversation patterns, CrewAI delegation, LangGraph state passing, and FIPA-ACL performatives. Teaches how to design what agents say to each other and in what order. Activate on: "agent conversation", "agent protocol", "multi-agent debate", "agent delegation", "supervisor worker pattern", "agent voting", "consensus protocol", "fan-out fan-in", "agent negotiation", "blackboard pattern", "agent dialogue", "conversation topology", "agent handoff". NOT for: wire format or serialization (use agent-interchange-formats), orchestration infrastructure (use agentic-infrastructure-2026), single agent behavior (use agentic-patterns).
The English summary is being prepared. Showing the original description.
Oracle-bound spec-review check that every structured Rollout prerequisite binds to an enforcement-type acceptance criterion. A prerequisite whose `enforced-by` is missing, dangling, or points at a non-enforcement AC is a BLOCKER (hard-blocks implementation entry via the checker fan-out verdict); a prerequisite left in free prose is ADVISORY. Read-only. Runs in the spec-review fan-out before implementation.
The English summary is being prepared. Showing the original description.
Run a governed R&D session — diamond 1 of the double diamond, which defines the problem and names which fan-out artifacts are warranted. Use when the operator pastes external material and says "consider this for gzkit", asks "is there something here gzkit is missing or could improve on", opens an exploratory design discussion whose outcome is not yet known, or asks why a class of defects keeps recurring and how the issues carrying it should be restructured. Ends at a six-row disposition map and operator sign-off; produces no fan-out artifact itself. Operator-invoked only.
The English summary is being prepared. Showing the original description.
Use when designing, running, or judging an experiment on a Syntropic137 workflow - comparing prompt variants, fan-out vs single agent, adversarial cross-model review, model selection (can Haiku do this job), phase splits, or tool grants. Trigger phrases include "run a workflow experiment", "A/B the workflow", "does fan-out help", "which model for this phase", "is this prompt better", "measure review quality", "why did the workflow do that". Do NOT use for a one-off workflow run that is just doing work, for authoring a workflow that is not being compared against anything, or for debugging a single failed execution (that is ordinary debugging).
The English summary is being prepared. Showing the original description.
GitHub Issueを分析し、1チケットの実装フロー（正本は /impl）へ fan-out する（複数Issue指定時は star 型で並列実行）。Triggers on: '/para-impl', '並列実装', 'Issueを実装して'
The English summary is being prepared. Showing the original description.
Authors reproducible bioinformatics pipelines with Snakemake - rules wired by output-file pattern, wildcards and expand() for sample fan-out, checkpoints for runtime-unknown outputs, resource/retry escalation, and conda/container software deployment on HPC and cloud. Use when deciding rule-based (Snakemake) vs channel/dataflow (Nextflow) authoring; wiring rules by OUTPUT-file pattern rather than imperative order; using wildcards + expand() for sample fan-out and constraining them to stop silent mis-routing; adding checkpoints when the set of outputs is unknown until a step runs (dynamic DAG); diagnosing why a job reran (or did not) under the mtime-plus-provenance trigger set; escalating memory on retry for OOM-killed jobs; and porting a Snakemake 7 `--cluster`/remote-provider command to the Snakemake 8+ executor-plugin and storage-plugin model (snakemake-executor-plugin-slurm) with `--software-deployment-method`.
The English summary is being prepared. Showing the original description.
Architects the encryption, codec/transport, and premium-feature layer of a group video-calling product, on top of an already-chosen SFU topology. Covers E2EE patterns (Zoom leader-fan-out, Signal pairwise-fan-out, MLS/TreeKEM as in Discord's DAVE, Telegram's ledger consensus), the SFrame (RFC 9605) + WebRTC Insertable Streams mechanism letting an SFU relay encrypted frames it can't read, codec/transport choice (AV1/VP9 SVC vs. simulcast, Opus, congestion control, Media-over-QUIC, WHIP/WHEP), and which "premium" features (virtual backgrounds, live captions, recording, breakout rooms, spatial audio) survive E2EE vs. structurally can't. Use when adding E2EE to an SFU-based call, choosing a 2026 codec/SVC strategy, or deciding which features to disable under E2EE mode. NOT for call topology/platform choice (see webrtc-adhoc-video-chat, which this composes with). NOT for one-way broadcast (see managed-video-streaming-pipeline). NOT for the DM/group-chat text backend (see realtime-messaging-backend-architecture).
The English summary is being prepared. Showing the original description.
Coordinate multiple AI agents with assigned owners, clear work boundaries, and a Kanban board that tracks progress. Organize separate branches or worktrees, collect test and handoff evidence, and define checks that must pass before combining their work.
Coordinate large migrations, reviews, or research tasks across multiple agents. Keep plans and intermediate results in scripts and files, challenge findings with independent checks, and integrate verified work in controlled batches.