The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill installs the eslint-plugin-lingui package as a development dependency. This is a standard vendor-provided tool for the Lingui i18n framework and is considered a legitimate resource for this workflow.
[COMMAND_EXECUTION]: The skill executes npx eslint to scan the codebase and npx lingui to manage translation catalogs. These are routine development commands executed within the project's local environment.
[PROMPT_INJECTION]: The skill ingests and processes strings extracted from the project's source code to decide whether they should be localized. This creates a surface for indirect prompt injection where malicious content within the audited codebase could attempt to influence the agent's logic. 1. Ingestion points: The agent reads /tmp/lingui-scan.json which contains source code snippets extracted via the ESLint scanner. 2. Boundary markers: No explicit delimiters or instructions are provided to the agent to treat the source content as untrusted data. 3. Capability inventory: The skill provides the ability to modify project files (wrapping strings in macros) and execute terminal commands. 4. Sanitization: There is no process described for sanitizing or filtering the source code content before it is reviewed by the agent.

find-unwrapped-strings