The Agent Skills Directory

[SAFE]: The skill instructions are focused entirely on business logic for pricing strategies and packaging. No malicious patterns such as credential harvesting, obfuscation, or unauthorized command execution were found.
[COMMAND_EXECUTION]: While the skill's configuration allows the use of Bash, the instructions do not include any shell commands that perform sensitive operations or bypass security controls. The environment is used solely for standard agent operations as defined by the author.
[EXTERNAL_DOWNLOADS]: The skill includes WebSearch and WebFetch in its allowed tools to gather market context. This is consistent with its stated purpose of researching pricing alternatives. No downloads of executable code from untrusted sources were identified.
[PROMPT_INJECTION]: The skill identifies a surface for indirect prompt injection due to its use of web tools. However, the use case is a strategy generator where data is used to inform business documents, presenting a negligible risk.
Ingestion points: WebSearch and WebFetch tools are used to gather competitor context (SKILL.md).
Boundary markers: No explicit delimiters for untrusted data are present.
Capability inventory: Access to Bash and Read (SKILL.md).
Sanitization: No explicit sanitization of fetched data is mentioned.

deal-pricing