Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
affaan-m avatar

Dynamic Workflow Mode

  • 1.1k installs
  • 238k repo stars
  • Updated August 5, 2026
  • affaan-m/everything-claude-code

dynamic-workflow-mode is a Claude Code skill that designs task-local agent harnesses, eval gates, and reusable skill extraction for adaptive agent workflows.

About

dynamic-workflow-mode is a Claude Code skill for designing task-local agent harnesses, eval gates, and reusable skill extraction. It helps a coding agent generate or adapt a custom loop, evaluator, crawler, or watcher for one task instead of only following a static command flow. A developer uses it when a task needs a custom harness, when multiple agents need the same repeatable process, or when work needs durable handoff artifacts and operator approval before merge. It matters because it turns ad-hoc agent scripting into a disciplined system with eval and merge gates.

  • Designs task-local agent harnesses with objective, inputs, outputs, eval, and handoff
  • Provides a decision tree for when to inline, build a harness, or extract a shared skill
  • Requires a task-specific eval gate before a workflow is called reusable

Dynamic Workflow Mode by the numbers

  • 1,116 all-time installs (skills.sh)
  • +217 installs in the week ending Aug 5, 2026 (Skillselion tracking)
  • Ranked #970 of 16,546 AI & Agent Building skills by installs in the Skillselion catalog
  • Data as of Aug 5, 2026 (Skillselion catalog sync)
At a glance

dynamic-workflow-mode capabilities & compatibility

Capabilities
agent harness design · workflow automation · eval gating · skill extraction
Use cases
orchestration · testing
From the docs

What dynamic-workflow-mode says it does

Design task-local harnesses, eval gates, and reusable skill extraction for Claude dynamic workflow mode and other adaptive agent harnesses.
SKILL.md
Do not claim a dynamic workflow is reusable until the eval can be rerun by another teammate.
SKILL.md
npx skills add https://github.com/affaan-m/everything-claude-code --skill dynamic-workflow-mode

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

Listed on Skillselion
Installs1.1k
repo stars238k
Last updatedAugust 5, 2026
Repositoryaffaan-m/everything-claude-code

What it does

Design a disciplined task-local agent harness with an eval gate, then decide when to extract it into a shared skill.

Who is it for?

Building a custom loop, evaluator, crawler, or watcher for a repeated task, or capturing a repeatable process as a shared skill.

Skip if: One-shot tasks, which should stay inline rather than inventing a harness.

When should I use this skill?

The user mentions dynamic workflows, custom harnesses, harness-per-task, or a task needs a custom loop, evaluator, or watcher.

What you get

A task-local harness with a pass/fail eval, control-pane checkpoints, and a handoff artifact, promoted to a shared skill only on repeat use.

  • a task-local harness or shared skill path
  • eval commands and results
  • a control-pane or handoff artifact

By the numbers

  • 5-branch harness decision tree
  • 5 harness contract elements (objective/inputs/outputs/eval/handoff)

Files

SKILL.mdMarkdownGitHub ↗

Dynamic Workflow Mode

Use this skill when a coding agent can generate or adapt a task-local harness instead of only following a static command flow. The goal is to turn dynamic workflow mode into a disciplined system: temporary harnesses for one-off work, shared skill extraction for repeated work, and observable control pane checkpoints for teams.

When To Activate

  • The user mentions dynamic workflows, custom harnesses, harness-per-task, adaptive workflows, or Claude Code dynamic workflow mode.
  • A task needs a custom loop, evaluator, crawler, fixture generator, watcher, or local dashboard.
  • Multiple agents need the same repeatable process but the process is not yet captured as a shared skill.
  • A workflow needs durable handoff artifacts, eval evidence, or operator approval before merge.

Core Contract

Dynamic workflow mode should produce a task-local harness only when the harness is cheaper and safer than manually driving the same steps. The harness must have:

  • Objective: the outcome it owns and the outcome it explicitly does not own.
  • Inputs: files, URLs, prompts, data sources, credentials policy, and user-provided constraints.
  • Outputs: commits, reports, screenshots, status files, or control pane snapshots.
  • Eval: at least one pass/fail check tied to the task, not only "it ran".
  • Handoff: a short artifact that tells the next operator what happened, what is blocked, and how to resume.

Dynamic Harness Decision Tree

1. One-shot task: keep it inline. Do not invent a harness. 2. Repeated task with changing inputs: create a task-local harness and keep it under a temp or project-local working area. 3. Repeated task across teammates or repos: extract the pattern into a shared skill. 4. Task with external state, queueing, or approvals: add control pane visibility before adding more automation. 5. Task with safety risk: add an eval gate and a human merge gate before autonomous execution.

Task-Local Harness Template

Use this structure before writing code:

# Dynamic Workflow Harness

Objective:
- Ship:
- Do not ship:

Inputs:
- Repo or workspace:
- External systems:
- Credentials policy:

Loop:
1. Discover current state.
2. Generate or update the smallest useful artifact.
3. Run eval checks.
4. Record status and handoff.
5. Stop on failed gate, unclear ownership, or unsafe external action.

Eval:
- Command:
- Expected pass signal:
- Failure owner:

Handoff:
- Status:
- Evidence:
- Next action:

Shared Skill Extraction

Promote a task-local harness into a shared skill only when at least two of these are true:

  • The same workflow appears in multiple sessions, repos, teams, or launches.
  • The workflow needs specific language, tool, or safety sequencing.
  • Failures repeat because operators skip a gate or lose context.
  • The workflow has a stable input/output contract.
  • The workflow benefits from a control pane, status board, or team handoff.

When extracting, write the skill first in skills/<name>/SKILL.md. Add command shims only if a legacy slash-entry surface is still required.

Control Pane Checkpoints

Dynamic workflow mode becomes team-usable when it exposes state. Record these checkpoints whenever the task spans more than one session:

  • Plan: objective, owner, acceptance criteria, and risky external systems.
  • Queue: work items, assigned agent role, branch/worktree, and dependency edges.
  • Run: active harness, current loop step, recent eval result, and token/cost signal if available.
  • Gate: test results, browser screenshots, security review, and merge readiness.
  • Handoff: what is done, what failed, what needs a human decision.

If the repo has ECC2 state enabled, prefer adding or reading checkpoints through the ECC control pane or state-store-backed scripts instead of scattering untracked notes.

Eval Gates

Every dynamic harness needs a task-specific eval. Pick the cheapest reliable gate:

Work TypeEval Gate
Code featureFocused test, lint, coverage, and one integration path
UI/control paneBrowser smoke with screenshot and overflow/error checks
Agent workflowFixture transcript or seeded work item with expected routing
Research/contentSource-neutral brief, claim checklist, and publish-ready outline
IntegrationDry-run command, config validation, and no-secret scan

Do not claim a dynamic workflow is reusable until the eval can be rerun by another teammate.

Anti-Patterns

  • Generating scripts that hide the real decision logic from the operator.
  • Treating dynamic workflow mode as permission to skip tests.
  • Creating one-off docs when a shared skill or status artifact is the real product.
  • Running multiple agents without ownership, merge gate, or conflict policy.
  • Letting raw private research data leak into public docs.

Output Standard

Finish with:

  • The harness or skill path.
  • The eval commands and results.
  • The control pane or handoff artifact path.
  • The next reusable extraction candidate.

Related skills

FAQ

When should I build a harness instead of working inline?

Keep one-shot tasks inline; build a task-local harness for repeated tasks with changing inputs, and extract a shared skill for tasks repeated across teammates or repos.

What must every harness have?

An objective, inputs, outputs, at least one pass/fail eval tied to the task, and a handoff artifact for the next operator.

AI & Agent Buildingagentsautomation

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.