The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill documentation provides instructions for installing standard developer tools from official and well-known registries. Examples include installing ruff and vulture via pip (PyPI), and goimports, gocyclo, golangci-lint, and staticcheck via go install from their respective official repositories on golang.org and github.com.
[COMMAND_EXECUTION]: The skill uses the Bash tool to perform repository analysis. It executes various CLI commands such as grep to find TODO comments, git blame to determine code age, and language-specific linters/formatters like ruff, vulture, and goimports to identify technical debt. All modifications in the 'Fix' phase are gated by explicit user consent.
[PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it ingests untrusted data from the repository's source code and comments.
Ingestion points: Reads all source files (.py, .go, .js, etc.) and specifically extracts text from TODO, FIXME, and HACK comments.
Boundary markers: The skill does not implement specific delimiters or 'ignore embedded instruction' markers when processing code content.
Capability inventory: The skill has Write and Bash capabilities, which are used to apply fixes and run tests.
Sanitization: There is no specific sanitization of the file content before it is included in the markdown report, although the skill's logic treats the data as text for reporting rather than executable instructions.

code-cleanup