The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Bash tool to retrieve technical context from the user's environment to fuel the brainstorming process.
Evidence: It executes git log --oneline -50 and history | grep -i error to find recent technical struggles and debugging sessions.
Context: These operations are restricted to local metadata retrieval and are consistent with the skill's stated purpose of 'problem mining' for content creation.
[PROMPT_INJECTION]: The skill is subject to indirect prompt injection risks as it ingest data from external files and command outputs.
Ingestion points: Files located in the content/ directory and the output of shell commands (git log, history) are read into the agent's context.
Boundary markers: The instructions do not specify explicit delimiters or 'ignore instructions' markers for the ingested text.
Capability inventory: The skill possesses Read, Write, and Bash capabilities, which could be abused if malicious instructions were embedded in the ingested content.
Sanitization: There is no evidence of sanitization or safety-filtering for the content read from the local environment.

topic-brainstormer