The Agent Skills Directory

[SAFE]: The skill performs legitimate reconnaissance of testing infrastructure. The use of Bash, Read, Glob, and Grep is scoped to identifying project configurations and test files (e.g., package.json, jest.config.ts, .github/workflows/).
[SAFE]: All external references, including the author domain (tonone.ai) and the repository (github.com/tonone-ai/tonone), are consistent with the vendor identity and do not represent a security risk.
[INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it ingests untrusted data from the repository's source code and configuration files.
Ingestion points: Project files read via Read, Bash, and Grep in SKILL.md (Steps 0-5).
Boundary markers: None explicitly defined in the prompt instructions to isolate external file content.
Capability inventory: Bash for shell execution, WebFetch/WebSearch for network access, and AskUserQuestion for user interaction.
Sanitization: No specific sanitization or validation of the file content is mentioned before processing.

proof-recon