The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by ingesting and analyzing potentially untrusted external data such as logs, events, and monitor descriptions.
Ingestion points: Data retrieved from Datadog via pup logs search, pup events search, and pup monitors list as specified in SKILL.md.
Boundary markers: Absent. There are no explicit markers or instructions to isolate analyzed data from the agent's core instructions.
Capability inventory: The skill has access to Bash, Write, and Edit tools, allowing it to execute commands and modify files based on its analysis.
Sanitization: Absent. The skill does not perform validation or sanitization of the data retrieved from Datadog before processing it.
[COMMAND_EXECUTION]: The skill uses the pup CLI (a tool from Datadog Labs) to perform its operations.
The allowed-tools section defines numerous Bash command patterns using the pup executable, including wildcard patterns like pup agent * and pup monitors search *, granting the agent broad interaction capabilities with the Datadog API.

datadog-analyser