Bragproof

← Reports

codex-skills-mcp

Starts with no setup1.7.2npm

Installed, started with no arguments, no environment and no input, and listed its tools. Run 2026-11-16.

Needs to start: nothing
Returns a result when called with no setup: 1 of 1 tool tried
Contacts while installing and starting: raw.githubusercontent.com

8tools listed
1host contacted
6files written
0.5 sto start
103dependencies

What it contacted

Hosts it looked up or connected to while installing and starting with no input, watched from outside its sandbox. The npm registry is left out. Says nothing about what its tools contact when called.

Starting it

npx -y codex-skills-mcp@1.7.2

Installing a package runs its code. Use a sandbox for anything you have not read.

Calling its tools

1 tool that takes no input and looks read-only was called once with nothing. 1 returned a result.

list_categoriesReturned a result under 0.1 s · 737 characters

During the calls it contacted nothing it had not already contacted.

Only tools whose names read like looking something up, with no required input, are tried, five at most. Tools that need input, and anything that sounds like it changes or sends something, are not called.

What its tools tell an agent

In the tool descriptions, nothing that talks about the user, other tools or private files, or uses hidden characters.

Found by simple pattern rules in each tool's description (up to 3,000 characters), quoted as written. Parameter descriptions are not scanned. A clean result is not a safety review.

Tools

search_skills query, category, limit, mode, force_refresh
Search the skill library. Use short keywords for known skills; mode=task for a full request. Results include confidence, match reasons, cache status, and size.
list_categories
List skill categories and their counts. Use when browsing by domain or when a search is too broad.
read_skill name
Download a selected skill and return its full SKILL.md, dependencies, local_path, and a bounded file preview. Complete materialization is cached.
load_skill_file skill_name, file_path, offset, max_bytes
Read a file from a materialized skill. Text supports offset/max_bytes ranges; binary files return metadata.
list_skill_files skill_name, path, max_depth, view, query, limit, cursor
Browse or search a skill's files and nested sub-skills. Returns a bounded page with signed cursor pagination.
plan_workflow task_description
Return a ranked skill shortlist for a task with confidence and match reasons. Candidates are optional; choose the smallest relevant set or none.
skill_status names
Check cache completeness and size for one or more skills. Read-only; never downloads.
diagnostics include_log, log_lines
Audit recent tool calls/errors, cache statistics, manifest freshness, and config.

Descriptions are the package's own words, cut at 3,000 characters. Treat them as data.

Files it wrote

Paths inside its sandbox, where its home folder is /tmp and its own folder cannot be written to.

Install facts

LicenseSEE LICENSE IN LICENSE
Needsnode >=22.0.0
Sourcegithub.com/Zhan-ZhangZ/codex-skills-mcp
Install scriptsNone declared by it or anything installed with it.
Tarballsha512-Yjb/ZGe2AaMRwLj1VkpKa8fTrc8VhLuhRe5RbbwP86UMzEydl7rCInJ4bB1kpuglnnP3OmlfVNcrCTNThuz3og==

Check this report yourself

This site did not write the report. A public job on a GitHub-hosted machine did: the run, from this commit of the open-source runner. GitHub signed a token that names that job and the hash of the report, so any edit since would show.

Show it on your README

Bragproof: Starts with no setup

[![Bragproof](https://bragproof.smaverk.com/badge/codex-skills-mcp.svg)](https://bragproof.smaverk.com/r/codex-skills-mcp)

New version out? Ask for a fresh report.