The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze a file named log.txt, which contains captured mitmproxy network traffic. This data source is external and untrusted, representing a potential surface for indirect prompt injection if the traffic logs contain adversarial instructions. However, the risk is minimized as the skill is instructional and does not provide the agent with autonomous tools to execute shell commands, perform network requests, or modify system states.
Ingestion points: Analyzes log.txt mentioned in the SKILL.md body.
Boundary markers: None are defined to separate the untrusted log data from the agent's instructions.
Capability inventory: No scripts, subprocess calls, or network operations are included in the skill.
Sanitization: No sanitization or validation logic for the input logs is provided.
[NO_CODE]: The skill consists entirely of Markdown content. It does not include any executable scripts, binary files, or automated shell commands. All suggested tools, such as mitmproxy and curl, are presented as examples for the user or the agent to use within its existing environment constraints.

mitm-find-callback