import-x-history
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill uses a vendor-managed API at
x.pcstyle.dev(and the aliasmdfromx.com) to retrieve X post data. Network access is restricted to these specific domains in the skill configuration. - [SAFE]: The skill correctly instructs the agent to handle API keys using environment variables (
XMD_FAST_KEY) and explicitly cautions against printing these secrets in the agent's output. - [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external user-generated content from X, which presents a surface for indirect prompt injection.
- Ingestion points: Post history is fetched via
curlrequests to thex.pcstyle.devAPI, as documented inSKILL.md. - Boundary markers: None are specified in the instructions; the skill does not explicitly prompt the agent to use delimiters or ignore instructions within the fetched post text.
- Capability inventory: The skill uses
Bash(curl)restricted to the vendor's infrastructure. - Sanitization: None; the skill provides raw text data from X posts which may contain untrusted strings.
Audit Metadata