plaud-transcripts-corrected-with-your-own-terminology-glossary

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

• Third-party content exposure detected (high risk: 0.75). The required workflow reads the user’s Plaud recording transcript (outsider-authored content from the Plaud source) and the glossary from Google Sheets, then uses those free-text artifacts directly in LLM prompts (“Correct the transcript against the user's glossary…” / “Produce the summary…”), creating an indirect prompt-injection path through Plaud-provided transcript text.