๐Ÿ”ฎ
AI EngineeringPython

Predict

by ffroliva

Predict is an AI Engineering skill for Claude Code, published by ffroliva in gflow-cli.

173 stars47 forkson ffroliva/gflow-cliAdded 2026/09/08+24% in starsRepository updated 2026/09/07
agenticai-agentsclaude-codecligenerative-aigoogle-flowimage-to-videoimagenmcpmcp-serverpythontext-to-videoveovideo-generation
Install in seconds
Install Predict
Copy Predict 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/ffroliva/gflow-cli/tree/develop/skills/predict ~/.claude/skills/predict

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

Without Node.js

git clone https://github.com/ffroliva/gflow-cli.git

Clones the whole repository, then copy the skillโ€™s own directory into your skills folder yourself.

In this catalog

Source file
skills/predict/SKILL.md in ffroliva/gflow-cli
Installs to
~/.claude/skills/predict
Collection
One of 18 skills cataloged from this repository
Category
AI Engineering โ€” 3670 skills

What Predict does

Predict runs a five-persona pre-implementation review for gflow-cli proposals before coding starts. Use it for high-stakes changes like auth, transports, selectors, schema migrations, or CLI surface changes to get a GO, CAUTION, or STOP verdict.

Predict is cataloged under AI Engineering on DirSkills. Predict comes from a repository tagged agentic, ai-agents, claude-code, cli and generative-ai.

Frequently asked about Predict

  • What else does ffroliva publish alongside Predict?

    Predict is one of 18 skills that DirSkills catalogs from ffroliva/gflow-cli, the repository it ships in. Its siblings there include Changelog, Check and Documentation Review. Each one is a separate skill with its own page in this directory, installs the same way Predict does, and is maintained by ffroliva in that same repository. The rest of the collection is listed on the ffroliva/gflow-cli page.

  • How does Predict compare to other AI Engineering skills?

    Predict ranks #3256 by stars among the 3670 AI Engineering skills in this catalog. The most-starred ones next to it are Architecture Decision Records, AI-First Engineering and Agentic OS. 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 Predict against them. Open each page to compare what they document and how they install.

More from ffroliva/gflow-cli

Predict is one of 18 skills cataloged on DirSkills from ffroliva/gflow-cli.

See all 18 skills โ†’
๐Ÿ“
1h ago

Changelog

Changelog shows unreleased changes and the last tagged version for quick comparison. Use it before a release or when you need a short view of queued and shipped work.
Writing
17347
โœ…
1h ago

Check

Check auto-fixes lint and formatting, then runs repo-wide verification for types, tests, doc links, and other quality gates before a commit. Use it when you want the codebase cleaned and checked against CI-style failures.
Quality
17347
๐Ÿ“š
1h ago

Documentation Review

Documentation Review checks release-facing docs for stale references, broken links, missing index entries, and drift between canonical and published pages. Use it before cutting a release or after major documentation changes.
Quality
17347
๐ŸŽฌ
1h ago

GFlow CLI

GFlow CLI automates Google Flow from the terminal or scripts for Veo video and Imagen image generation. It is used for text-to-video, image-to-video, image pipelines, batch runs, and other Flow credit workflows through a headed Chrome session.
Automation
17347
๐Ÿงช
1h ago

Issue Assessment

Issue Assessment reviews a gflow-cli issue against code, tests, and docs to decide whether the claim is confirmed, needs more info, or is already known. It also checks whether the issue can be verified end to end and drafts a reporter-facing reply.
Quality
17347
๐Ÿ› ๏ธ
1h ago

Issue Resolve

Issue Resolve drives an assessed gflow-cli bug to a draft PR in an isolated worktree. It is used for localized, verifiable fixes with test-first changes and browser-free validation, or to note when human e2e verification is still needed.
Automation
17347