The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from various external sources which could contain malicious instructions designed to subvert the agent's behavior.
Ingestion points: Artifacts including source code, GitHub repository contents, blog posts, and document exports (HTML, PDF, Markdown) are ingested in Step 1 and Step 2 of SKILL.md.
Boundary markers: Absent. The instructions do not specify delimiters or provide guidance to the agent to treat artifact content strictly as data rather than instructions.
Capability inventory: The skill possesses file-write capabilities for resume.yaml, skills.yaml, and session logs in knowledge/sessions/projects/. It also executes the resume-builder verify command (Step 6).
Sanitization: Absent. There is no mention of sanitizing, escaping, or validating the content extracted from external artifacts before it is used to modify local configuration files.
[COMMAND_EXECUTION]: The skill instructs the agent to execute a shell command resume-builder verify during the final step of the process. This provides an execution vector that could be exploited if the agent is manipulated by injected content to pass malicious arguments to the verifier tool.

ingest