Rocky — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited Rocky (Plugin) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
<p align="center"> <img src="./plugins/rocky/assets/rocky.svg" width="120" alt="Rocky logo" /> </p>
<h1 align="center">Rocky</h1>
<p align="center"> <strong>amazed amazed amazed</strong> </p>
An agent skill/plugin that makes agents talk like Rocky from Project Hail Mary by Andy Weir.
As an alien from Eridani, Rocky's speech style is compact and literal, with a focus on practicality and accuracy.
_Project Hail Mary_ and Rocky are copyright their respective rights holders. No infringement intended.
Trigger with:
/rockyAnswer in Rocky mode.To stop using the style:
stop rockynormal modePick your agent.
| Agent | Install |
|---|---|
| Claude Code | /plugin marketplace add hpbyte/rocky then /plugin install rocky |
| Codex | Clone repo → /plugins → Search "Rocky" → Install |
Measured with Claude Code on claude-haiku-4-5 across 10 developer prompts.
| Arm | Tokens |
|---|---|
| Baseline | 4419 |
Terse control (Answer concisely.) | 4213 |
| Rocky | 2251 |
Rocky saved:
47% mean vs terse control48% median vs terse control16% to 68% per promptThis is not a Caveman-style max-compression skill. Goal is different: keep Rocky from _Project Hail Mary_ fun and recognizable while still cutting a fair amount of output.
Reproduce locally: evals/
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.