The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides the ability to execute arbitrary shell commands via the slot command and pre-configured repository check and rebuild scripts. This is the primary function of the tool for orchestrating development workflows.
Evidence: scripts/land.py (slot command handling), scripts/landing_lib/_governor/build.py (run_build function), and scripts/landing_lib/_queue/integration.py (rebuild and check methods).
[DYNAMIC_EXECUTION]: The main script scripts/land.py uses a custom module loading system that manipulates sys.modules and uses types.ModuleType to dynamically load its internal libraries from relative paths. This is implemented for namespace isolation.
Evidence: Initial initialization block in scripts/land.py.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from git commit logs and pull request descriptions. This data is processed and used to populate PR titles and bodies, which creates a potential surface for indirect prompt injection if an agent interprets this metadata as instructions.
Evidence: Ingestion points in scripts/landing_lib/store.py (pr_body function) and capability inventory involving subprocess execution and GitHub CLI operations. No LLM-specific sanitization or boundary markers are employed for this external content.

landing