The Agent Skills Directory

[DATA_EXFILTRATION]: The skill explicitly targets and reads sensitive environment configuration files including .env, .env.local, and .env.production. These files are high-value targets as they typically contain API keys, database credentials, and other secrets. Although the instructions restrict the agent from echoing values, the underlying tool has full access to the file contents.
[COMMAND_EXECUTION]: The skill executes a local script (scripts/envlint.py) using Python 3 via the Bash tool. The security of the skill depends entirely on the integrity of this script, which is not provided for analysis but is granted access to the environment's secrets.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection by processing untrusted data from local environment files.
Ingestion points: Reads content from .env, .env.example, and other variable files.
Boundary markers: None present. The skill relies on natural language role-play instructions ("act as an env-var auditor") to maintain safety boundaries rather than technical delimiters.
Capability inventory: Access to the Bash tool and the ability to read arbitrary files in the project directory.
Sanitization: No sanitization or validation of the ingested file content is mentioned, allowing potentially malicious instructions embedded in keys or values to be processed by the model.

env-lint