The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructions require the agent to ingest and process various user-controlled files, creating a vulnerability surface where malicious instructions embedded in the textbook content could influence agent behavior.
Ingestion points: The skill reads from docs/course-description.md, docs/learning-graph/03-concept-dependencies.csv, docs/glossary.md, and all Markdown files under docs/**/*.md (referenced in Step 1 and Step 2 of SKILL.md).
Boundary markers: Absent. There are no instructions for the agent to use XML tags, triple quotes, or specific warnings to ignore instructions embedded within the source content during analysis.
Capability inventory: The skill has file-read access to the entire docs directory and file-write access to create docs/faq.md, logs/faqs.md, and various report files in docs/learning-graph/. It also has the capability to modify the project configuration file mkdocs.yml (referenced in Step 10 of SKILL.md).
Sanitization: Absent. The workflow does not include steps to sanitize or filter the content read from external files before using it to generate the FAQ or update the configuration.

faq-generator