The Agent Skills Directory

[COMMAND_EXECUTION]: The skill relies on executing system commands to perform its core functions.
The scripts/ensure-ledger Bash script executes command -v ledger, ledger version, and ledger capabilities to verify readiness.
scripts/ledger_context.py uses subprocess.run to call git for workspace discovery and the ledger binary for storage integrity checks (ledger doctor).
scripts/ensure_custody.py invokes the ledger binary to perform validations and transactions required for data migration.
[EXTERNAL_DOWNLOADS]: The skill manages the lifecycle of its native dependency through a package manager.
scripts/ensure-ledger installs or upgrades the tkersey/tap/ledger Homebrew formula, which is a vendor-owned resource.
[DYNAMIC_EXECUTION]: The migration logic involves dynamic loading of Python code from sibling skills.
scripts/ensure_custody.py uses importlib.util to load and execute 'importer' modules defined in the migration.json files of other skills located in the same installation root. This is a modular design for extending migration support to different data types.
[INDIRECT_PROMPT_INJECTION]: The skill processes external workspace data and migration definitions, presenting a surface for indirect instructions.
Ingestion points: Reads migration.json declarations and inventories regular files and symlinks within the user's workspace.
Boundary markers: Employs structured JSON envelopes and explicit schema versioning (e.g., ledger-owner-migration/v1) to delimit data.
Capability inventory: The skill can execute shell commands, perform filesystem mutations (archiving, copying), and load dynamic modules.
Sanitization: Path inputs are strictly validated via a relative() helper that prevents absolute paths, null bytes, or directory traversal (..).

ledger