The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell commands to interact with the local environment and manage repository state.
Evidence: Executes git submodule update --remote --merge in Phase 1.1.
Evidence: Uses find and grep to scan for files and analyze their contents in Phases 1.1, 7.1, and 7.3.
[EXTERNAL_DOWNLOADS]: The skill initiates network requests to remote git repositories.
Evidence: The git submodule update command in Phase 1.1 fetches updates from remote sources defined in the project's submodule configuration.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process and synthesize content from external, potentially untrusted submodule files, creating a vulnerability surface for indirect injection.
Ingestion points: The skill reads SKILL.md and other markdown files from paths under submodules/*/ during the discovery phase (Phase 1.1).
Boundary markers: No explicit boundary markers or 'ignore' instructions are used when interpolating external content into the merging process.
Capability inventory: The agent has the capability to execute shell commands (git, find, grep) and write new or modified skill files to the local filesystem (Phase 6).
Sanitization: No sanitization, escaping, or security validation of the ingested external content is described; the focus is on quality and semantic similarity rather than security integrity.

syncing-submodules