mcpField
S
skills-vote-local

skills-vote-local

github.com/memtensorAgent workflows30669 installsUpdated 12w ago
Repository

About skills-vote-local

A skill for discovering skills in a local or private library rather than through a network service. It routes the request through a script and provides instructions for main-agent, subagent, or fallback operation.

Key features

  1. 01Local and private skill discovery
  2. 02Prompt routing through a script
  3. 03Subagent route when configured
  4. 04Fallback route when delegation is unavailable

Use cases

  1. 01Find skills in a private library
  2. 02Route a lookup through a configured subagent
  3. 03Use fallback when subagent execution is unavailable

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/memtensor/skills-vote --skill skills-vote-local

After installing, type/skills-vote-localin Claude Code to confirm it loaded.

Bundled files

References and scripts installed along with SKILL.md.

configs/
config.yaml.example
docs/
config-schema.md
handoff.md
scripts/
check_env.py
chroma_utils.py
cli_common.py
config_validation.py
index.py
local_skill_search.py
prompt.py
recommend.py
route_prompt.py
sync_skills.py

Details

From the project README

Skills Vote Local

Read only this file first. Do not read scripts/, docs/, or any other files in this skill unless this file or a script output explicitly tells you to do so.

Use this skill when the current task needs skill discovery from a local or private skill library instead of a network service.

All paths mentioned in this file are relative to this skill root. cd to this root directory before running any command here.

Run:

uv run -qq scripts/route_prompt.py --role main

Then follow the rendered stdout exactly.

If routing.mode selects a subagent route, the user configuration explicitly requests subagent-based skill lookup. The main agent should use subagent delegation when the current host/tooling permits it.

Do not use fallback merely because the current user message did not repeat "use subagent".

If the current host, tool policy, runtime, or authorization model cannot create a subagent in this turn, run:

uv run -qq scripts/route_prompt.py --role main --fallback

Then follow the fallback stdout exactly.

If you are a subagent and you read this file directly, do not run this command with --role main. Run:

uv run -qq scripts/route_prompt.py --role subagent

Then follow the rendered stdout exactly.

Comments

Markdown supported