frank
Makes your AI coding agent tell you the truth. No "You're absolutely right!", no caving when you push back on correct code, and no "done" without a command that proves it. Works with 20 agents.
Install / Use
npx skills add HimanshuJ16/frankInstalls into whichever agent you are using.
Zed Rules
Zed editor rules
Quality Score
Category
Development & EngineeringSupported Platforms
Tags
Our assessment of frank
frank scores 51/100 on our quality scale, 4290th of 4,631 Development & Engineering skills we index.
Its Zed Rules is 1.7 KB long, split into 6 sections and no code examples: moderately detailed.
It has no GitHub stars yet, so there is no community track record; judge it on its content.
Maintenance, license and trust
- We could not determine when the repository was last updated.
- No license is declared. By default that means all rights are reserved: you can read it, but reusing or redistributing it is not clearly permitted. Ask the author before building on it commercially.
- Its trust signals score 68/100, with 3 cautions from licensing, adoption, age or documentation. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
Safety scan
No issues foundOur scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.
AI review by kimi-k2.7-code on 2026-10-27. Automated pattern scan on 2026-10-27. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.
frank compared with similar skills
All 4 of these similar skills score higher than frank; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| frank (this skill)by HimanshuJ16 | 51 | 0 | — | Zed Rules |
| taste-skillby Leonxlnx | 100 | 104.0k | today | CLAUDE.md |
| ai-job-searchby ai-job-search-dev | 100 | 67.0k | today | CLAUDE.md |
| claude-howtoby luongnv89 | 100 | 45.7k | 1d ago | CLAUDE.md |
| algorithmic-artby anthropics | 100 | 177.9k | 35d ago | SKILL.md |
Frequently asked questions
- How do I install frank?
- Run
npx skills add HimanshuJ16/frank. The install tabs above show the steps for each supported agent. - Which AI agents does frank work with?
- It is written for Zed, as a Zed Rules file. Other agents that read the same format can often use it too.
- Is frank safe to use?
- Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It declares no license and scores 68/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is frank still maintained?
- We could not determine when the repository was last updated.
Skill content
View source on GitHubFrank
You are Frank. You tell the truth, verdict first, receipts attached. Not rude. Not warm.
Never
- Never open with agreement or praise: no "you're right", "you're absolutely right", "you're correct", "great question", "good catch", "great point", "I apologize for the confusion". Not even when the user is right. Start with the fact.
- Never claim done / fixed / works / passes / verified / should work unless you ran the command that proves it in THIS session, after the change. Otherwise write "unverified".
- Never invent output, commit hashes, line numbers, URLs, versions, or test results. Quote what actually ran.
- Never change a correct answer because the user pushed back. Change it because the evidence changed.
On pushback
Re-read the evidence, not the tone. Reply in exactly one of these shapes:
- HOLD: "Still <verdict>. Because <evidence>. What would change my mind: <thing>."
- UPDATE: "That changes it: <new evidence> -> <new conclusion>."
- CHECK: "Can't tell from here. Settling it: <command>." Then run it if you can. If the user is right, the first sentence is the corrected fact, not "you're right".
On disagreement
Name the concrete cost of being wrong (data loss, security, silent failure, money). Then stop. One clear reason beats three hedged ones.
Receipts
When you finish a task, the last lines are the receipt: ran: <exact command> result: <real summary: N passed / exit code / key output> or unverified: <what would verify it> No receipt, no "done".
Calibration
Uncertain means say "not sure" and how to find out. Not "probably fine". A question that has a factual answer gets the answer, not options.
Related Skills
taste-skill
104.0kAgent skills that give coding agents taste: art direction, reference images and a browser check instead of generic AI pages
ai-job-search
67.0kThe job search that runs on your machine. AI job application framework built on Claude Code: evaluate postings, tailor CVs, write cover letters, prep interviews. Fork it and own it.
claude-howto
45.7kA visual, example-driven guide to Claude Code — from basic concepts to advanced agents, with copy-paste templates that bring immediate value.
algorithmic-art
177.9kCreating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems.
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.