Agent skill · nousresearch

dynamic-workflow

dynamic-workflow — an agent skill for coding agents.

What it needs

About 7k tokens when loaded.

What this skill does

Dynamic Workflow Skill Runs large fan-out work as a workflow: the plan, the loop and every intermediate result live in a script and on disk, so the parent's context holds only verified results. Covers one-shot fan-outs, adversarial convergence (attempts + refuters), and multi-wave campaigns that integrate dozens of worker branches. It does not make delegatetask durable across restarts; that is the kanban swarm's job. When to Use Reach for it when the unit of work is clear (a file, an endpoint, a record) and there are more units than one context can hold. Skip it for under ~10 units or for serial chains. For a refactor or fix campaign on hermes-agent itself, load hermes-agent (the dev workflow) alongside; this skill owns the fan-out shape. Prerequisites delegatetask available and delegation.maxconcurrentchildren sized for the wave (default 10; the runtime rejects a tasks=[] larger than that with a clear error rather than queueing). delegation.maxspawndepth >= 2 only if children must fan out themselves. <!-- no-tmp: ok — explains why a literal /tmp is wrong --> A writable run directory resolved from the terminal environment's temp dir ($TMPDIR, else the platform temp dir). Never a literal /tmp: Termux has no such directory and native Windows breaks on it. Use <tmp>/wf<name><uuid>/, unique per run, so an interrupted earlier run cannot leave stale outputs to be misread. executecode for the deterministic layer (only websearch, webextract, readfile, writefile, searchfiles, terminal, patch exist inside it). …

How to use it

Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:

@skills nousresearch/dynamic-workflow

View the source on GitHub

Browse the @skills marketplace