sysadmin-toolbox

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill explicitly auto-refreshes its reference files from the public upstream GitHub repo (see "Keeping Current" pointing to scripts/refresh.sh which pulls from github.com/trimstray/the-book-of-secret-knowledge) and the SKILL.md workflow states the agent will load and consult those reference files (e.g., references/shell-oneliners.md) as part of its actions, so untrusted third‑party content can be ingested and materially influence tool recommendations and next actions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly auto-refreshes its reference files from the upstream GitHub repo at runtime via ./scripts/refresh.sh ("References auto-refresh weekly ... from the upstream repo"), so content fetched from https://github.com/trimstray/the-book-of-secret-knowledge is retrieved at runtime and directly controls the agent's reference/prompts used to generate responses.