The Agent Skills Directory

[COMMAND_EXECUTION]: The skill is designed to execute local CLI tools and scripts as part of its database management workflow. For instance, the activation instructions include a fallback mechanism to execute a local Node.js script: node .aiox-core/development/scripts/unified-activation-pipeline.js data-engineer. It also utilizes standard utilities such as git, psql, pg_dump, and supabase-cli to perform project status checks and database operations.
[DYNAMIC_EXECUTION]: The agent dynamically constructs shell commands based on the environment and user input. The coderabbit integration determines its execution path based on the host operating system, specifically handling paths for macOS/Linux (~/.local/bin/coderabbit) and Windows (wsl bash). Furthermore, the run-sql command permits the execution of arbitrary SQL strings provided by the user or from local files.
[INDIRECT_PROMPT_INJECTION]: The skill identifies a vulnerability surface for indirect prompt injection through the processing of local project files.
Ingestion points: The agent ingests data from project stories in docs/stories/, handoff artifacts in .aiox/handoffs/, and user-specified SQL files via the apply-migration and run-sql commands.
Boundary markers: There are no explicit markers or 'ignore' instructions defined for the content read from handoff artifacts or story files.
Capability inventory: The agent has the ability to execute shell commands, run Node.js scripts, and perform database operations (DDL/DML) via SQL tools.
Sanitization: The skill contains explicit instructions to redact sensitive credentials automatically and to validate user input before constructing dynamic SQL queries.

aiox-data-engineer