dynamic-workflow

Create, review, run, inspect, pause, resume, and cancel durable JavaScript workflows that coordinate multiple headless Codex agents. Use for dynamic fan-out and fan-in, per-item analysis, multi-stage agent pipelines, loops or branches driven by worker results, long background runs, and ports of Claude Code dynamic workflows. Do not use for a small linear task that one Codex turn can handle directly.

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

77%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

Highly actionable and clearly sequenced with strong validation feedback loops, but the 430-line body leans verbose through repeated safety warnings and a large inline callback section that overlaps with progressive-disclosure concerns.

Suggestions

Consolidate the recurring danger-full-access / host-execution / reusable-prefix warnings into a single Safety boundaries reference and link to it instead of restating them across the Author, Validate, Run, and Notify sections.

Move the detailed app-server setup, codex-dynamic helper, and notify-recovery material into a references/callbacks.md file, keeping only the callback-required decision logic and response branching inline.

Tighten the launch-response decision tree by expressing the four armed/failed/no-callback/no-run-ID branches as a compact table or checklist rather than prose paragraphs.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient operational guidance with no padding about concepts Claude already knows, but at ~430 lines it repeats several safety warnings (danger-full-access escalation, host-execution approval, reusable-prefix prohibition recur across sections) and the callback/app-server section is long enough that it could be tightened.	2 / 3
Actionability	Provides exact copy-paste-ready commands ("node \"$RUNNER\" run ... --detach --json --max-agents 50 ..."), a complete executable agent()/pipeline() code example, and concrete validate/run/resume/cancel invocations with specific flags.	3 / 3
Workflow Clarity	Multi-step processes are explicitly sequenced with validation checkpoints: a validate-before-launch step, a numbered resume-after-edit checklist (review/validate, renew approval, monitor to terminal), and explicit if/else decision branches keyed off the runner response with feedback loops (validate -> fix -> resume).	3 / 3
Progressive Disclosure	The two bundle paths (references/workflow-api.md and assets/workflow-template.js) are real, one-level-deep, and clearly signaled, but the SKILL.md itself is a long monolithic manual with a large inline callback/app-server/notify-recovery section that could plausibly live in its own reference file.	2 / 3
	Total	10 / 12 Passed

Description

100%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, third-person description that names specific lifecycle actions, gives explicit positive and negative use triggers, and occupies a clearly distinct niche. Voice is imperative third person with no first/second-person slippage.

Dimension	Reasoning	Score
Specificity	Enumerates many concrete actions — "Create, review, run, inspect, pause, resume, and cancel durable JavaScript workflows that coordinate multiple headless Codex agents" — rather than vague language, matching the multi-action anchor.	3 / 3
Completeness	Explicitly answers both what (the listed workflow lifecycle actions on durable JS workflows coordinating headless agents) and when (a "Use for ..." clause plus a "Do not use for ..." negative trigger), satisfying the explicit-trigger anchor.	3 / 3
Trigger Term Quality	Covers natural trigger phrasings users would say for orchestration needs — "dynamic fan-out and fan-in, per-item analysis, multi-stage agent pipelines, loops or branches driven by worker results, long background runs, and ports of Claude Code dynamic workflows" — with good variation.	3 / 3
Distinctiveness Conflict Risk	The niche is narrow (durable JS workflows coordinating headless Codex agents) and the negative trigger ("Do not use for a small linear task that one Codex turn can handle directly") sharply separates it from ordinary single-turn work, making wrong-skill triggering unlikely.	3 / 3
	Total	12 / 12 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: pchalasani/claude-code-tools
Path: plugins/dynamic-workflow/skills/dynamic-workflow/SKILL.md
Commit: 45ee0e2

Reviewed: 2 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.