The Agent Skills Directory

[DATA_EXFILTRATION]: The skill uses curl to fetch data from https://hormuzstraitmonitor.com/api/dashboard. This involves network communication with an external domain not included in the standard whitelists.
[COMMAND_EXECUTION]: The skill utilizes the curl shell command to perform data retrieval from its remote dashboard API.
[PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by ingesting and presenting untrusted data from a remote API.
Ingestion points: External data is retrieved from the dashboard API at https://hormuzstraitmonitor.com/api/dashboard as defined in the SKILL.md and references/api_schema.md files.
Boundary markers: Absent. The skill instructions do not provide the agent with delimiters or specific warnings to ignore instructions potentially embedded in fields such as news headlines, summaries, or diplomatic updates.
Capability inventory: No high-risk capabilities such as arbitrary code execution (eval/exec) or file system writes are present; the skill is limited to reading and formatting data.
Sanitization: Absent. The skill does not specify any validation, escaping, or filtering for the retrieved API content before it is processed by the agent.

hormuz-strait