The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: Fetches official error code definitions from Ably's documentation site at https://ably.com/docs/platform/errors/codes to validate error hints.
[EXTERNAL_DOWNLOADS]: Instructs the environment to install the typescript-language-server and typescript packages from the npm registry if they are not already present.
[COMMAND_EXECUTION]: Executes shell commands including git diff and git log to identify changes, and npm install for setting up required language server dependencies.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted data in the form of branch code changes, which represents a potential attack surface for indirect prompt injection.
Ingestion points: Modified source code files, git diff outputs, and git log history (SKILL.md Step 1 and Step 3).
Boundary markers: None specified to prevent the agent from following instructions embedded within the code it is reviewing.
Capability inventory: npm install (Step 0), git commands (Step 1), WebFetch for external URLs (Step 3), and LSP structural analysis (Step 3).
Sanitization: No explicit sanitization or filtering is performed on the code content before it is processed by the agent.

ably-review