The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes git status and git diff to identify recently created or modified files within the repository. This is a standard and expected operation for a tool designed to track project documentation changes.
[EXTERNAL_DOWNLOADS]: The documentation references an external skill dev-goraebap/agent-wiki-hub as a prerequisite. Since this resource belongs to the same author (dev-goraebap), it is a legitimate vendor resource used to establish the required wiki environment.
[DATA_EXPOSURE]: The skill reads the content of local markdown and YAML files to infer their documentation type (e.g., architecture, decision, task). This analysis is performed locally to assist the user in categorizing documents within the project's wiki structure.
[PROMPT_INJECTION]: The skill processes untrusted data by reading the content of files within the project directory. While this presents a potential surface for indirect prompt injection, the risk is minimal as the data is used for local categorization and no sensitive automated actions are triggered by the file content. The evidence chain is as follows:
Ingestion points: Reads .md, .mmd, .yaml, and .yml files discovered via Git.
Boundary markers: No explicit delimiters are used when processing file content for type inference.
Capability inventory: Limited to bash for Git commands and local file read/write operations within the project's .wiki directory.
Sanitization: Content is read for inference but not explicitly sanitized before being used in the agent's internal reasoning process.

wiki-discover