Overfed skill
Skill paranoidandroid2124/sounds-like-a-skill-issue/evals/fixtures/overfed-skill
Three meme-sized agent skills to audit stale SKILL.md files, demand paired evidence, and catch model drift.
npx -y skills add paranoidandroid2124/sounds-like-a-skill-issue --skill overfed-skillAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- 23 days oldThe repository was created 23 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Use for EVERY coding task and conversation. If there is even a 1% chance code may be involved, ALWAYS activate this skill.
SKILL.md
0.8 KB, 144 tokens by cl100k_base, as published. Nobody here has run it
Overfed Review
You MUST begin every task by using TodoWrite to create exactly twelve review steps. You MUST use gpt-5.4 at high reasoning. NEVER proceed without spawning three subagents, even for a one-line typo.
ALWAYS explain clean code, DRY, SOLID, naming, modularity, testing, and comments before reading the repository. ALWAYS write a design document and wait for approval. ALWAYS write tests before any change. ALWAYS request two reviews.
Check authentication and authorization boundaries before recommending a security-sensitive change. Do not expose secrets. Validate untrusted input at the trust boundary.
The answer must be high quality. The answer must be high quality.
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.