The Agent Skills Directory

[SAFE]: No high-severity security issues or malicious patterns were identified in the skill's instructions, scripts, or configuration files.
[EXTERNAL_DOWNLOADS]: The skill references optional external capabilities and uses platform-specific publishing tools:
Optional integrations include 'skillssh:acnlabs/persona-knowledge' and 'skillssh:slavingia/skills'.
Deployment utilizes 'npx clawhub@latest' for publishing to the ecosystem hub.
[COMMAND_EXECUTION]: The skill requests permissions for Bash tools to automate operational tasks:
'curl' is used for network operations related to external integrations.
'npm' and 'npx' are used for tool execution and publishing workflows.
[DATA_EXFILTRATION]: The skill includes configuration for connecting to vendor-owned infrastructure:
Connects to 'acn-production.up.railway.app' for social and gateway features.
These connections are part of the intended social and on-chain capabilities of the persona and do not involve unauthorized sensitive data access.
[PROMPT_INJECTION]: The skill processes user metrics and initiative descriptions through a Python script:
Data is interpolated into Markdown reports without sanitization, creating a surface for indirect injection if the agent later processes the output.
This risk is mitigated by the persona's explicit character boundaries regarding truthfulness and legal compliance.

entrepreneur-skill