upwork-client-messenger
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process "thread context" pasted from external freelance client conversations, which represents an untrusted data source.
- Ingestion points: Input field for "Thread context, pasted" in
SKILL.md. - Boundary markers: Absent; there are no instructions to the agent to treat the pasted content as data only or to ignore embedded instructions.
- Capability inventory: The skill is limited to text drafting and does not appear to utilize tools for file writing, network operations, or command execution.
- Sanitization: Absent; no sanitization or validation of the pasted input is performed.
- [SAFE]: The skill references official resources and documentation from Upwork, which is a well-known service in the freelance industry.
- [SAFE]: The skill includes a shortcut to read identity and rate information from a local configuration file (
upwork-profile.md) located in the working directory or the~/.claude/directory to provide personalized message drafts.
Audit Metadata