From agent-almanac
Authors a reusable, deterministic agent orchestration workflow as a self-contained .mjs script for Claude Code's Workflow tool, with structured phases, fan-out, and verification.
How this skill is triggered — by the user, by Claude, or both
Slash command
/agent-almanac:create-workflowThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Author a workflow — the fifth content type — as a self-contained `workflows/<name>.mjs` script run by Claude Code's Workflow tool. A workflow fixes its phases, fan-out, and verification structure in JavaScript: its **control flow** is deterministic and rereadable, while the **outputs** of its `agent()` calls are LLM subagents and remain nondeterministic. This skill is the *procedure* for author...
Author a workflow — the fifth content type — as a self-contained workflows/<name>.mjs script run by Claude Code's Workflow tool. A workflow fixes its phases, fan-out, and verification structure in JavaScript: its control flow is deterministic and rereadable, while the outputs of its agent() calls are LLM subagents and remain nondeterministic. This skill is the procedure for authoring one; guides/creating-workflows.md is the API reference it draws on.
Vendor-API caveat. The Workflow run model is generally available on paid Claude Code plans (~v2.1.154+), but the script-authoring surface (the injected
agent()/parallel()/pipeline()/phase()/log()/workflow()primitives and theargs/budgetglobals) is an evolving vendor API — accurate as observed in Claude Code v2.1.x, subject to change, never CI-enforced.
workflows/ library, or authoring a personal workflow for .claude/workflows/.review-changes). This becomes the filename stem, the sidecar name:, and meta.name — all three must be identical.pipeline() vs barrier parallel()).args (default them so it runs with no input).A workflow is one of five content types. Choose deliberately:
Expected: A one-line justification for why this is a workflow and not a team or a single skill.
On failure: If the coordination must adapt turn-by-turn, stop and use create-team. If only one agent acts, use create-skill.
Start from the canonical scaffold — never a blank file:
cp workflows/_template.mjs workflows/<name>.mjs
For a personal (non-library) workflow, copy into .claude/workflows/<name>.mjs instead. The template carries the sidecar block, a pure-literal meta, a phase(), a pipeline() fan-out, an agent({ schema }) call, and the hard constraints inline.
Expected: A new .mjs file that is a verbatim copy of the template.
On failure: If workflows/_template.mjs is missing, you are on a pre-Phase-1 checkout — fetch main.
Rename in the three places that must stay identical — the filename stem, the sidecar // name:, and the meta.name literal:
grep -n "name:" workflows/<name>.mjs # sidecar + meta.name must equal the filename stem
That triple equality (filename ↔ sidecar name ↔ meta.name) is the Workflow({ name }) and /<name> discovery contract.
Expected: All three read the same kebab-case name.
On failure: A Workflow not found by name error at runtime means the triple is out of sync — re-check all three.
meta and Sidecarexport const meta must be a pure literal — no variables, function calls, spreads, or template interpolation. Required fields: name and description. phases (one entry per phase the workflow uses, whether opened by a global phase() call or a stage's per-call phase: option) is optional but recommended. Mirror name, description, and the phases titles in the top-of-file sidecar comment block — the sidecar is the catalog source of truth (the analogue of YAML frontmatter on the other content types), readable by grep without a JS parser.
Expected: meta is a literal object; the sidecar agrees with it; phases ⊇ every title passed to phase() or a stage phase:.
On failure: A meta must be a pure literal error means a value references a variable or call — inline it.
The body runs inside an async wrapper — use top-level await and a top-level return directly. Compose with the injected globals (no imports):
pipeline(items, ...stages) — the default: each item flows through every stage with no barrier between stages (wall-clock = slowest single chain).parallel(thunks) — a barrier: use only when a stage needs every prior result at once (dedup, an early-exit count, a synthesis step).agent(prompt, { schema, label, phase, agentType }) — spawn one subagent; with { schema } it returns a validated object.phase(title), log(message), and the args / budget globals. Default args so the workflow runs with no input.Pass a JSON Schema as { schema } to force structured output (no free-text parsing). See guides/creating-workflows.md for the full primitive reference.
Expected: A body that defaults its inputs, fans out with the right primitive, and returns a value.
On failure: If you reach for parallel() only to flatten or map between stages, that barrier is not justified — do the transform inside a pipeline() stage.
agent({ agentType }) names the spawn type per call — the workflow's native expression of the persona-vs-spawn decoupling. The intent rule applies:
isolation: 'worktree') must target an implementing agent type.Explore).Set agentType explicitly on every stage so the contract is visible, not implied.
Expected: Every agent() call names an agentType whose capability matches what the stage does.
On failure: A mutating stage targeting an advisory type cannot write — switch it to an implementing type such as general-purpose.
If your workflow verifies candidate findings (the classify → refute → synthesize spine), bake in the three lessons the review-changes seed encodes:
agent() returns null when a subagent is skipped or dies, so filter(Boolean) before counting and require a majority of confirmations to survive (confirmedVotes >= Math.floor(n / 2) + 1). Surviving when "few enough refuted" inverts the fail-safe — a dead refuter would let an unverified finding through.Expected: Survival gates on a confirmation quorum, verifiers share the proposer's evidence, and null results are filtered.
On failure: If real findings vanish, check that verifiers aren't starved of context (lesson 2); if junk survives, check the gate counts confirmations, not refutations (lesson 1).
Workflow scripts use a top-level return, which the runtime accepts (it wraps the body in an async function) but raw ESM rejects — so plain node --check reports Illegal return statement on a valid workflow. Use the wrap-then-check recipe:
{ echo '(async()=>{'; \
sed 's/^[[:space:]]*export const meta/const meta/' workflows/<name>.mjs; \
echo '})()'; } | node --check -
The authoritative check is running it: Workflow({ name: '<name>' }).
Expected: The wrap-check passes with no syntax error.
On failure: Do not "fix" a top-level return to pass raw node --check — that would cripple the script. Use the wrap-check; debug any other syntax error normally.
The Workflow tool resolves Workflow({ name }) from .claude/workflows/<name>.mjs. In Phase 1, install by hand:
cp workflows/<name>.mjs .claude/workflows/<name>.mjs # curated installs prefix: almanac-<name>.mjs
Then invoke Workflow({ name: '<name>' }) or the /<name> slash command. .claude/workflows/ is user-writable and the save-flow writes there, so a curated install must namespace (almanac-<name>.mjs) to avoid shadowing a user's own workflow.
Expected: The workflow runs end-to-end and returns its value.
On failure: If /<name> is not found, confirm the file is in .claude/workflows/ and the triple-name contract holds (Step 3).
Phase-2-pending. A
workflows/_registry.yml, CI validation of the sidecar, and a CLI install adapter are deferred behind the #288 promotion gate (~8–10 workflows + a real install request). Until they land, a library workflow needs no registry entry — the sidecar frontmatter is its catalog metadata and discovery is by filename. When the registry exists, this step becomes "add an entry derived from the sidecar."Workflows are excluded from i18n. Unlike skills/agents/teams/guides, a workflow is executable code, not prose — do not scaffold translations for it.
If contributing a reviewed seed to agent-almanac, place it in workflows/, cross-reference it from guides/creating-workflows.md, and carry the vendor-API caveat in any prose you add.
Expected: A library workflow lives in workflows/ with an accurate sidecar; no registry or translation steps are attempted in Phase 1.
On failure: If a tool expects workflows/_registry.yml, you are ahead of the promotion gate — stop and confirm Phase 2 has shipped.
workflows/<name>.mjs (or .claude/workflows/<name>.mjs for personal use).name:, and meta.name are identical (triple-name contract).export const meta is a pure literal; sidecar mirrors name/description/phases.phases: ⊇ every title passed to phase() or a stage phase: option.args, uses an appropriate fan-out primitive, and returns a value.agent() call sets an agentType whose capability matches the stage (advisory vs implementing).filter(Boolean) null results.Date.now(), Math.random(), argless new Date(); no TypeScript syntax; no filesystem/Node APIs.node --check recipe passes.workflows/_registry.yml entry or translation scaffold was created (both are Phase 2 / i18n-excluded).null/dead refuter pass an unverified finding through. Gate on a majority of affirmative confirmations.meta. export const meta must be a literal — no spreads, calls, or interpolation. A computed meta fails at load.return. It is valid Workflow dialect; rewriting it to satisfy raw node --check breaks the script. Use the wrap-check.Date.now() / Math.random() / argless new Date() break workflow resume. Pass timestamps via args; vary randomness by agent index or label.workflows/_registry.yml or scaffold translations for a workflow — registries/CLI/validation are gated, and workflows are i18n-excluded.create-skill — author a SKILL.md (the how for one agent); the meta-pattern sibling.create-agent — define an agent persona; workflows spawn agents by agentType.create-team — compose a declarative, model-driven roster (the adaptive counterpart to a workflow).commit-changes — commit the new workflow file.review-codebase / review-pull-request — the single-reviewer skills the review-changes workflow coordinates many of.npx claudepluginhub pjt222/agent-almanacCreates multi-agent orchestration workflows from natural language using Socratic questioning, pattern detection for sequential/parallel/conditional flows, temp script creation, and syntax generation with visualization.
Orchestrates deterministic JS workflows for Claude Code subagents with phases, parallelism, and quality patterns. For fan-out to hundreds of agents or codebase-wide audits.
Validates, runs, and debugs multi-agent YAML workflows. Use when orchestrating AI agents, configuring routing, or setting up human-in-the-loop gates.