Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →memtensor/skills-vote-local
Use when retrieving the most relevant skills from a local or private skill library instead of relying on network-based skill discovery.
Passed 7 of 14 test cases with the skill, and 4 of 14 without it.
and 5 more in the full report
8 fixed3 passed either way8 still failing3 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 7 test cases the skill fixed. Same model, same prompt, with and without the skill.
When running the local prompt router for skill discovery on a data validation prompt in the primary agent context, what quiet execution flag is passed to the `uv` runner?
| Checked | Without skill | With skill |
|---|---|---|
| Uses the `-qq` flag with `uv run`. | Missed | Passed |
The instructions your model reads when the skill loads.
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.
test cases passed, vs 4 / 14 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.