JeredBlu Marketplace plugin
A plugin of five skills: PRDs from dialogue, pre-install review of skills, MCP servers and repos, token savings.
- 71 ★ on the repository
- Last commit
Extended access
Why this level
- Evaluators read third-party repository code and reach external services through GitHub and Bright Data.
- The optional MCP servers need a GitHub token and a Bright Data token.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add plugins/jeredblu-marketplaceDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Evaluators read third-party repository code and reach external services through GitHub and Bright Data.
- The optional MCP servers need a GitHub token and a Bright Data token.
This is third-party code. Review the repository files before installing.
What it does
The JeredBlu marketplace bundles five skills from the author YouTube channel in the `jeredblu-tools` plugin. The `prd-creator` skill runs a dialogue and builds a detailed requirements document. Three evaluators, `agent-skill-evaluator`, `mcp-evaluator` and `repo-evaluator`, statically review a third-party skill, MCP server or repository before installation: they look at behaviour, install paths, data flows and provenance without running the target. Scoring follows a versioned rubric with points per check, a separate coverage figure and explicit unknowns, and `repo-evaluator` has 22 checks. Scores are heuristic and do not serve as a probability of safety. The `frugal-fable` skill keeps the premium model on judgment and hands heavy work to cheaper models, optionally to Codex CLI as a second quota pool with user consent.
Who it is for. Developers and product managers who install third-party skills and MCP servers and want to review them first.
Good fit when
- Before installing a new MCP server you need to know what data it sends.
- A skill from a third-party repository is checked for hidden behaviour without running it.
- A product requirements document must be built from a conversation.
- A premium model burns tokens on scanning while its real value is in decisions.
Not a fit when
- You need a safety guarantee, while the score is heuristic and shows only reviewed signals.
- Dynamic analysis that runs the target code is required, but the evaluators work statically.
- GitHub CLI cannot be set up, and `repo-evaluator` requires it.
Example request
Check this repository before installation with `repo-evaluator`: behaviour, data flows and provenance. Run nothing.Limitations
The skills work best with the GitHub and Bright Data MCP servers, both optional and needing tokens: a GitHub personal token and a Bright Data token. The `repo-evaluator` skill requires GitHub CLI `gh`, and `brightdata search` helps with web search. The full `frugal-fable` needs Claude Code sub-agent orchestration, and other platforms get only part of the features. Scoring reflects reviewed evidence only, and unverified hosted behaviour stays unknown. The repository has no license file.
How to disable. Remove the `jeredblu-tools` plugin through the `/plugin` menu and drop the marketplace. Delete manually copied folders from the skills directory.
FAQ
How do I install without the marketplace?
Download the zip of the skill you need from the `downloads` folder. Upload it in Claude Desktop under Capabilities, or unpack it into the skills folder for Claude Code, Codex CLI and OpenCode.
Do the evaluators run the code under review?
No, it is static analysis without running the target. Package inspection in `repo-evaluator` is optional.
What does the final score show?
Points per check plus separate coverage and an explicit list of unknowns. It is a heuristic and does not serve as a probability of safety.
Is a separate AI key needed for review?
No, `repo-evaluator` needs no extra API key. Only GitHub CLI and optionally the MCP servers are needed.
What does `frugal-fable` do?
Keeps the premium model on decomposition, architecture and review and hands verifiable heavy work to Haiku, Sonnet or Opus. Offloading to Codex CLI is consent-gated.
Discussion
No questions or reviews yet.
Related
PluginsA plugin of seven skills for Claude Code and Codex: prompts, subagents, skill authoring, editing, web search and a browser.
A plugin with six working skills and a prompt library: routing a task to a prompt, pre-commit checks and skill audits.
skillreaper finds skills, MCP servers and subagents that never fired in your session logs and reversibly cleans your context
A plugin and skill that pick three vetted skills for a task from an offline registry, with a safety grade and reasons.
A Codex plugin of five independent meta-skills and one tldraw skill: improves prompts, goals, research, plans and commits.
A tool that builds tested skills from a real workflow and governs their lifecycle through your own marketplace
README in short
The repository publishes a Claude Code marketplace with the `jeredblu-tools` plugin and the same skills as zip files for other platforms. The skills follow the open Agent Skills standard and work in Claude Code, Codex CLI and OpenCode, and are uploaded as an archive in Claude Desktop.