x-thread-builder
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data, which creates an attack surface for indirect prompt injection.
- Ingestion points: The skill gathers "raw material" (notes, transcripts, results) from the user in Step 1 of
SKILL.mdto be used as the basis for the thread. - Boundary markers: There are no instructions or patterns provided to the agent to use delimiters or specific boundary markers to isolate this untrusted user input from the skill's core instructions.
- Capability inventory: The skill includes a functional capability to publish content to an external platform (X/Twitter) via the
lib.publishtool call described in Step 10 ofSKILL.md. - Sanitization: The instructions lack specific validation, filtering, or sanitization routines to ensure that the ingested raw material does not contain instructions that could override the agent's intended behavior.
Audit Metadata