Workflow router

Skill alainlebret/claude-agents/higher-ed-teaching-agents/skills/workflow-router

Selects the best workflow-spec dynamically from the repository, computes the minimal skill chain, and emits a deterministic routing decision with confidence and fallback. Use when the orchestrator receives a pedagogical request and must choose among multiple workflows without hardcoded chains.From its SKILL.md

Install

npx -y skills add alainlebret/claude-agents --skill workflow-router

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.0 KB, 353 tokens by cl100k_base, as published. Nobody here has run it

Workflow Router

Goal

Map an instructor request to exactly one workflow in workflows/*/workflow-spec.json, with explicit rationale and fallback behavior.

Inputs

Instructor request (free text)
Optional mission.json
Repository workflow inventory under workflows/

Produce

Write routing-decision.json with:

request_summary
candidate_workflows (ranked)
selected_workflow
selected_reason
confidence (high|medium|low)
required_skills
optional_skills_selected
skip_conditions_applied
human_confirmation_required (bool)
fallback_workflow

Selection policy

Discover candidates by reading all workflow-spec.json files.
Score candidates on:

intent match (create, redesign, evaluate, validate, benchmark);
artifact intent (exam/lab/course/project/grading/figures);
constraints (language_mode, evaluation_mode, level, duration, toolchain);
required input availability.

Select the highest score only if score gap to second candidate is significant.
If ambiguity remains, set human_confirmation_required: true.

Skill-chain policy

Start from workflow execution_order.
Apply skip_conditions if artifacts already exist and pass validators.
Add optional skills only when their trigger in optional_extensions is satisfied.
Never invent a skill not present in workflow spec.

Rules

Do not use hardcoded workflow names in logic; load from repository each run.
Emit one selected workflow, never multiple.
If confidence is low, provide a fallback and require confirmation before generation.
Keep routing deterministic for same inputs and same repository state.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Gives 0 of the 12 instructions most automation workflows skills give in 353 tokens

Counted across 745 of the 1,008 authors here whose files we hold, read 2026-08-07

Write conventional commit messagesin 36 of 745, across 35 files
Delete branches after mergein 30 of 745, across 21 files
Make atomic commitsin 25 of 745, across 15 files
Write minimal code to pass testsin 22 of 745, across 10 files
Re-snapshot after navigation or DOM changesin 21 of 745, across 13 files
Use try-catch for error handlingin 20 of 745, across 8 files
Run tests before committingin 20 of 745, across 12 files
Write tests before implementationin 20 of 745, across 8 files
Configure branch protection rulesin 19 of 745, across 5 files
Explain the why in commit messagesin 19 of 745, across 9 files
Refactor code while tests remain greenin 19 of 745, across 6 files
Interact with elements using refsin 19 of 745, across 11 files

Said here and by no other author read

map request to one workflow
read all workflow spec files
score candidates on intent and constraints
select highest scoring candidate
require significant score gap before selecting
set human confirmation flag if ambiguous

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.