The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes git and gh (GitHub CLI) to perform repository inspection, view pull request metadata, and manage the commit workflow. These operations are standard for a development-focused skill and use locally anchored configuration such as the origin's fetch URL.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input in the form of code diffs and repository files, creating a potential surface for instructions embedded in code to influence the agent.
Ingestion points: The agent reads code changes and repository state using git diff, git status, and gh pr view (SKILL.md, Scope section).
Boundary markers: The instructions do not specify the use of delimiters or 'ignore' markers to separate untrusted code content from the agent's instructions.
Capability inventory: The skill allows for file modification, committing changes, and executing shell commands for verification (SKILL.md, Process section).
Sanitization: There is no explicit requirement for the agent to sanitize or validate the content of the code it reads before processing.
[DYNAMIC_EXECUTION]: The skill requires the agent to execute the project's test suite and type-checkers to verify that its 'deslop' operations have not changed application behavior. While this involves running executable code, it is restricted to the local project's existing testing infrastructure and serves a defensive verification purpose.

deslop