The Agent Skills Directory

[NO_CODE] (HIGH): The core logic of the skill resides in the scripts/ and references/ directories, which are completely missing. This makes it impossible to verify the safety of fullstack_scaffolder.py, project_scaffolder.py, or code_quality_analyzer.py before they are executed on the host system.
[PROMPT_INJECTION] (HIGH): The skill is susceptible to Indirect Prompt Injection (Category 8) due to its design for analyzing external codebases. \n
Ingestion points: The skill reads user-supplied project paths and files for analysis (scripts/project_scaffolder.py <target-path>). \n
Boundary markers: None. There are no instructions to the agent to treat content within the scanned directories as untrusted or to ignore embedded instructions in analyzed code. \n
Capability inventory: The skill allows for python script execution, environment variable manipulation (cp .env.example .env), and infrastructure deployment (docker build, kubectl apply). \n
Sanitization: No sanitization or safety-wrapping logic is described for handling external file content.
[COMMAND_EXECUTION] (HIGH): The skill facilitates the execution of arbitrary Python scripts provided within the skill's own directory and shell commands for deployment and package management. Without the script contents, these represent an unverified execution risk.
[EXTERNAL_DOWNLOADS] (MEDIUM): Instructions include npm install and pip install -r requirements.txt. These commands pull code from external registries (npm, PyPI) without provided lockfiles or manifests, posing a risk of dependency confusion or supply chain attacks.

senior-fullstack