twmd-supporters
Sync Portaly donation notification emails (Gmail) into the supporters SSOT via canonical SUPPORTERS-PIPELINE — checkpoint from last_fetched, read full email bodies (never snippets, since amount/monthly-type can be truncated), dry-run before write, regen the two privacy-partitioned derived views for /about#sponsors, verify no cross-layer privacy leak before commit. TRIGGER when: user says "抓贊助", "更新贊助者", "跑 Portaly", "supporters sync", or routine `twmd-supporters-weekly` fires.
Low
Low-risk findings.
1 low severity finding. Worth noting, but not necessarily harmful.
W011: Third-party content exposure detected (indirect prompt injection risk).
What this means
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
Why it was flagged
In SKILL.md the required runtime workflow explicitly performs a Gmail search and then reads FULL_CONTENT of each email (“Stage 2 Gmail search + 逐封讀 FULL_CONTENT”) which is outsider-authored free text from senders, creating an indirect prompt-injection exposure path.
- Repository
- frank890417/taiwan-md
- Commit
4d66426
- Audited
- Security analysis
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.