investor-sourcing
Pass
Audited by Gen Agent Trust Hub on Jul 30, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches an upstream configuration file from the vendor's domain at
https://monid.ai/SKILL.mdand installs the@monid-ai/clipackage from the official npm registry. - [COMMAND_EXECUTION]: The skill executes shell commands using the
monidCLI to fetch social media data and manage local snapshots. It also provides instructions for users to manually set upcronorlaunchdfor task scheduling if desired. - [PROMPT_INJECTION]: The skill processes untrusted external data, which presents a surface for indirect prompt injection.
- Ingestion points: External data is ingested from X (Twitter) user profiles, followings, and search timelines, as well as from company websites via the
WebFetchtool. - Boundary markers: The skill does not implement specific delimiters or 'ignore' instructions to separate untrusted data from the agent's internal logic.
- Capability inventory: The agent has the capability to execute shell commands via the
monidCLI and write files to the local filesystem (e.g.,snapshots/anddigests/). - Sanitization: No explicit sanitization or filtering of the fetched social media or web content is performed before the data is processed by the model.
Audit Metadata