The Agent Skills Directory

[DATA_EXFILTRATION]: The skill accesses sensitive configuration files including .env, .env.*, and fnox.toml to audit and migrate secret management to 1Password. It identifies key names for migration but explicitly instructs the agent never to print, commit, or paste actual secret values.
[COMMAND_EXECUTION]: The skill executes shell commands using the mise tool manager to install packages and verify project health via mise trust, mise install, and mise run. These commands are run within the context of the target repository to align it with project standards.
[EXTERNAL_DOWNLOADS]: Fetches project standards and best practices from the The-Focus-AI/standards repository. It also installs CLI tools and agent skills from the npm registry and other remote sources using mise use and skills add.
[PROMPT_INJECTION]: The skill processes untrusted data from the target repository's documentation and agent instructions, creating a surface for indirect prompt injection.
Ingestion points: Reads AGENTS.md, CLAUDE.md, .cursor/rules, and README files from the target repository.
Boundary markers: No explicit delimiters or instructions are used to separate untrusted file content from the agent's instructions.
Capability inventory: The skill possesses the ability to execute shell commands (mise, fnox, skills) and write to the local filesystem.
Sanitization: There is no evidence of sanitization or filtering of the content ingested from the target repository.

standardize-project