The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill facilitates the download of external resources required for Zed extension development. Assets such as the Makefile grammar are fetched from the well-known tree-sitter-grammars organization on GitHub, and official JSON schemas are downloaded from zed.dev. The example extensions demonstrate standard patterns for downloading language server binaries from developer repositories (e.g., artempyanykh/marksman) and installing npm packages from the @modelcontextprotocol organization.
[COMMAND_EXECUTION]: The development workflow involves executing local commands to build and verify extensions. The assets/examples/verify.sh script runs cargo build, cargo test, and cargo clippy to ensure Rust crates are correctly compiled. Additionally, the skill utilizes bundled Python tools to validate extension metadata, syntax queries, and themes against established rules.
[INDIRECT_PROMPT_INJECTION]: The skill processes extension source files provided by developers, presenting a surface for potential indirect prompt injection.
Ingestion points: The agent reads extension.toml, Rust source files, and Tree-sitter query files (.scm) from the extension directory.
Boundary markers: No explicit delimiters or warnings are used when processing these files.
Capability inventory: Subprocess execution for building (Cargo) and static analysis (Python-based checkers).
Sanitization: The bundled scripts (check_extension.py, etc.) perform validation of the input file structures and values against expected schemas and allowed keys.

develop-zed-editor-extensions