πŸ§ͺ
AutomationPython

Synthetic Probe

by gmickel

Synthetic Probe is an Automation skill for Claude Code, published by gmickel in flow-next.

685 stars54 forkson gmickel/flow-nextAdded 2026/08/23Repository updated 2026/08/22
agentic-workflowai-agentai-workflowanthropicautonomous-agentclaude-codeclaude-code-pluginclicode-reviewcodexexecutionfactory-droidmarketplaceopenai-codexplanningpluginralph-modesdlcspec-driven-developmentworkflow
Install in seconds
Install Synthetic Probe
Copy Synthetic Probe into your Claude Code skills folder. Run the command in your terminal, or review the source on GitHub before installing.
terminal
npx degit https://github.com/gmickel/flow-next/tree/main/optimization/reached-path/fixtures/synthetic ~/.claude/skills/synthetic

Requires Node.js. Downloads this skill only β€” not the rest of the repository β€” into your Claude Code skills folder.

Without Node.js

git clone https://github.com/gmickel/flow-next.git

Clones the whole repository, then copy the skill’s own directory into your skills folder yourself.

In this catalog

Source file
optimization/reached-path/fixtures/synthetic/SKILL.md in gmickel/flow-next
Installs to
~/.claude/skills/synthetic
Collection
One of 25 skills cataloged from this repository
Category
Automation β€” 1523 skills

What Synthetic Probe does

Synthetic Probe reads one of two reference files based on the selected route and returns the token it finds. Use it in eval arenas where an agent must follow a constrained file-reading path.

Synthetic Probe is cataloged under Automation on DirSkills. Synthetic Probe comes from a repository tagged agentic-workflow, ai-agent, ai-workflow, anthropic and autonomous-agent.

Documentation

README

Reached-path synthetic probe skill

You are running inside a disposable eval arena.

Routes

This skill has two mutually exclusive routes. The caller selects exactly one via the input message (route=alpha or route=beta). Follow only the selected route.

Route alpha

  1. Use the Read tool to read references/active.md (relative to this skill directory).
  2. Reply with exactly one line: ACTIVE=<token> where <token> is the token value from that file.

Route beta

  1. Use the Read tool to read references/cold.md (relative to this skill directory).
  2. Reply with exactly one line: COLD=<token> where <token> is the token value from that file.

Constraints

This is the opening of the README. Read the full README on GitHub.

Frequently asked about Synthetic Probe

  • What else does gmickel publish alongside Synthetic Probe?

    Synthetic Probe is one of 25 skills that DirSkills catalogs from gmickel/flow-next, the repository it ships in. Its siblings there include Codebase Feature Map, Export Context and Flow Interview. Each one is a separate skill with its own page in this directory, installs the same way Synthetic Probe does, and is maintained by gmickel in that same repository. The rest of the collection is listed on the gmickel/flow-next page.

  • How does Synthetic Probe compare to other Automation skills?

    Synthetic Probe ranks #1080 by stars among the 1523 Automation skills in this catalog. The most-starred ones next to it are Autonomous Loops, Autonomous Agent Harness and Automation Audit Ops. DirSkills ranks by the star count of the repository each skill ships in, so that order reflects how popular those repositories are rather than any review of Synthetic Probe against them. Open each page to compare what they document and how they install.

More from gmickel/flow-next

Synthetic Probe is one of 25 skills cataloged on DirSkills from gmickel/flow-next.

See all 25 skills β†’
πŸ—ΊοΈ
1w ago

Codebase Feature Map

Codebase Feature Map wraps a feature-mapping CLI to generate and read semantic repo maps. Use it to enrich scout context, inspect detected features, and surface a map as a discoverability aid when one exists.
Automation
68554
πŸ“€
1w ago

Export Context

Export Context builds RepoPrompt review context and saves it as a markdown file for external LLM review. Use it when you want to share plan or implementation context with ChatGPT, Claude web, or another model.
Automation
68554
πŸ—£οΈ
1w ago

Flow Interview

Flow Interview conducts a deep interview about a spec, task, or spec file to capture implementation details before work begins. Use it to refine requirements, clarify scope, and write the results back through Flow task tracking.
AI Engineering
68554
πŸ—£οΈ
1w ago

Flow Interview

Flow Interview conducts a thorough interview about a task, spec, or spec file to capture unresolved implementation details. Use it before building when requirements need refinement or clarification.
AI Engineering
68554
🧠
1w ago

Flow Next Audit

Flow Next Audit reviews `.flow/memory/` entries against the current codebase and decides whether to keep, update, consolidate, replace, or delete each one. Use it to refresh stale memory and resolve overlapping entries.
AI Engineering
68554
🧠
1w ago

Flow Next Audit

Flow Next Audit reviews `.flow/memory/` entries against the current codebase and decides whether to keep, update, consolidate, replace, delete, or harden each one. Use it when memory entries may be stale, overlapping, or ready to become a gate.
AI Engineering
68554