Loading...
Loading...
Found 10 Skills
Only to be triggered by explicit super-swarm-spark commands.
Analyze a task, pick the right fleet type, and generate a ready-to-launch fleet (fleet.json + prompt.md files). Discovers available fleet skills dynamically. Use when the user wants to run work in parallel, asks to "plan a fleet", or says "fleet-plan".
Spawn isolated agents for parallel task execution. Local mode auto-selects Codex sub-agents or Claude teams. Distributed mode uses tmux + Agent Mail (process isolation, persistence). Triggers: "swarm", "spawn agents", "parallel work".
Use when the user says "vm", "voice mode", "team", "coordinate", or needs to orchestrate multiple agents working on related tasks in parallel
Only to be triggered by explicit super-swarm commands.
Orchestrate parallel implementation with coder/overseer pairs. Coders implement decomposed tasks using evanflow-tdd; overseers review each coder's output for bugs, gaps, errors, AND cohesion violations against a shared contract. A final integration overseer checks cross-coder cohesion. Use for plans with 3+ truly independent tasks that share an interface contract.
Orchestrate parallel subtasks across multiple repositories using vibe-kanban MCP. Use this skill when: (1) Work needs to be distributed across multiple repos, (2) Parallel execution of independent tasks is needed, (3) You need to create subtasks and collect their results, (4) Multi-agent coordination via kanban task descriptions. Triggers: "subtask", "multi-repo", "parallel tasks", "distribute work", "orchestrate".
Launch OpenAI Codex CLI as a subagent (ChatGPT subscription auth, no API key). Use when delegating a self-contained coding task to Codex from another agent — parallel implementation work, a second opinion, or an independent verification pass.
Single-agent recursion and parallel fan-out within one AG2 `Agent` — auto-injected `run_subtask` / `run_subtasks(parallel=True)` (opt in via `tasks=TaskConfig(...)`) for self-delegation, and `Agent.as_tool()` as a lightweight no-hub way to call one named agent from inside another. Use when a single coordinator wants to break work into its own sub-tasks, fan out concurrent sub-tasks, or invoke a specialist agent as a tool. Covers context flow, recursion safety, and `persistent_stream` for sub-task history. **For two or more agents actually collaborating with a registry, durable channels, governance, or turn-taking, use `ag2-network-quickstart` instead** — the network is the standard multi-agent pattern in AG2.
STOP — if you were dispatched as a subagent to execute a briefed task, this skill is not for you: it governs the dispatching orchestrator only, and the mode and path rules you need are already in your brief. Do the task you were briefed with instead. Use whenever the user asks to dispatch, launch, spawn, fan out, or delegate work to an agent, subagent, worker, or coding agent in a separate git worktree — phrasing like "dispatch a subagent", "launch an agent", "fan out subagents", "parallel agents", or "use a separate worktree to implement/fix/refactor/test/review X" — or whenever two or more workers must edit the same repository concurrently without colliding. Symptoms it prevents: worktrees created from the wrong base commit, workers accidentally editing the orchestrator checkout when an isolated worktree was intended (Mode D is the sanctioned single-writer exception), Edit/Write failing outside the agent CWD, colliding diffs, shared Docker/e2e infrastructure clobbered by concurrent test runs, merged worktrees left consuming disk, and force-removed worktrees losing uncommitted work.