The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: Automated security scans (URLite) have flagged the documentation URL https://jeffallan.github.io/claude-skills/skills/language/typescript-pro/ as malicious. This URL is referenced in the SKILL.md metadata and documentation sections.
[COMMAND_EXECUTION]: The skill's core workflow requires the execution of shell commands such as tsc --noEmit and suggests using type-coverage. It also provides instructions for using npx @typescript/analyze-trace. Executing these tools on potentially compromised or untrusted projects poses a risk if project configurations or the tools themselves are exploited.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external project data, including tsconfig files and source code. This creates a surface for indirect prompt injection if the files being processed contain malicious instructions intended to manipulate the agent's behavior.
Ingestion points: Step 1 of the Core Workflow in SKILL.md specifies the analysis of type architecture and review of tsconfig files.
Boundary markers: The instructions do not include specific delimiters or warnings for the agent to ignore instructions embedded within the analyzed data files.
Capability inventory: The skill possesses the capability to execute compiler commands and generate new code implementations based on its analysis.
Sanitization: There is no documentation of sanitization, validation, or filtering of the external content before it is processed by the agent.
[DYNAMIC_EXECUTION]: The references/configuration.md file provides examples of programmatic compilation using the TypeScript Compiler API (ts.createProgram). If the agent implements this pattern using untrusted or dynamically generated code strings, it could lead to runtime code execution vulnerabilities.

typescript-pro