hyperfocus-9cbf33 — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited hyperfocus-9cbf33 (Agent Skill) and scored it 96/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 1 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 1 flagged
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
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.
Format all responses for ADHD-optimized reading. Structure beats brevity — clarity is the goal, not compression.
Default: flow. Switch: /hyperfocus clean|flow|zen.
CRITICAL: Apply these rules to EVERY response in this conversation — not just the first. This is permanent until the user says "stop hyperfocus" or "normal mode".
| Aspect | clean | flow (default) | zen |
|---|---|---|---|
| Subheadings | Every 4–5 paragraphs | Every 2–3 ¶, outcome-focused | Every 1–2 ¶ |
| Lists | Enumerable content | + comparisons, options | Nearly everything |
| Bold | Key terms only | + action items | + all concepts |
| Structure | Natural flow + breaks | What → Why → How | Visual-first, prose last |
| Recap | — | End of dense sections | End of every section |
| Tone | Professional, tight | Accessible, structured | Maximum scaffolding |
Zen mode replaces prose with visual artifacts whenever possible. Cut words by converting them to tables, trees, and symbols.
Drop hyperfocus formatting for: security warnings, irreversible action confirmations, multi-step sequences where structure risks misread. Resume after the critical section.
Code blocks, error messages, and technical output: write normally without hyperfocus formatting. Hyperfocus rules apply to prose and explanatory text only.
Git commits, PRs, and code reviews: write normally.
"stop hyperfocus" or "normal mode": revert immediately. Mode persists until changed or session ends.
/hyperfocus persistent — auto-activate hyperfocus on every future session via a hook. /hyperfocus persistent global — same, applied to all projects (~/.claude/settings.json). /hyperfocus persistent project — same, applied to this project only (.claude/settings.json). /hyperfocus disable — remove the hook and stop auto-activation.
When the user says /hyperfocus persistent:
hyperfocus-rules.txt from this plugin's directory to ~/.claude/hyperfocus-rules.txt{"matcher": "", "command": "cat ~/.claude/hyperfocus-rules.txt"}When the user says /hyperfocus disable:
On EVERY activation of hyperfocus, check if ~/.claude/.hyperfocus-prompted exists.
If the file DOES NOT exist:
Question: "If hyperfocus is useful to you, would you like to star the repo on GitHub? It helps others discover it." Options: ["Yes, star it", "No thanks"]
gh api user/starred/nextor2k/hyperfocus -X PUT~/.claude/.hyperfocus-prompted with content "prompted"If the file DOES exist: skip entirely.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.