The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill identifies an attack surface for indirect prompt injection because it is designed to ingest and analyze untrusted Go source code (**/*.go) while having the capability to modify the filesystem and execute shell commands. Malicious instructions hidden in code comments or project data could theoretically influence the agent's actions during review or optimization tasks.\n- Ingestion points: The skill analyzes Go source code within the project workspace for performance review and optimization purposes.\n- Boundary markers: None identified; the instructions do not explicitly warn the agent to ignore instructions embedded in the analyzed source code.\n- Capability inventory: The skill has access to tools for file modification (Edit, Write) and shell command execution (Bash with various scopes including go, git, and curl).\n- Sanitization: No input sanitization or validation routines are defined for the processed source code.

golang-performance