The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of user-controlled strings (such as the bet name or area) and incorporates them into a shell command argument.
Ingestion points: User-supplied input from the argument-hint ('bet name or area') and potential file content referenced via @file syntax.
Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions to prevent embedded commands in the user input from affecting the shell execution.
Capability inventory: The agent is instructed to execute a local Python script (tools/allocate-id.py) and perform file write operations across multiple directories (strategy/, context/).
Sanitization: There is no requirement or guidance for the agent to sanitize or escape the user-provided 'short reason' before passing it to the --note argument of the command line.
[COMMAND_EXECUTION]: The skill requires the execution of a workspace-local script, tools/allocate-id.py. While this is a functional requirement for ID management, the execution of shell commands based on user-provided parameters is a known attack surface that relies on the agent's internal safety guardrails for mitigation.

strategic-bet