The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or security risks were detected in the skill analysis. The skill is instructional in nature and does not request or utilize any system tools or network access.
[PROMPT_INJECTION]: The skill uses detailed instructional prompting to define its persona as a digital literacy expert. No attempts to bypass safety filters, extract system prompts, or override agent constraints were found. Natural instructional language is used appropriately for the educational context.
[DATA_EXFILTRATION]: The skill does not access sensitive file paths (e.g., credentials, SSH keys) nor does it perform any network operations. There are no hardcoded secrets or credentials.
[REMOTE_CODE_EXECUTION]: There are no remote downloads, script executions, or package installations (npm/pip) present in the skill.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an ingestion surface for untrusted data which is interpolated into prompts.
Ingestion points: source_type, evaluation_context, student_level, specific_source, student_profiles, subject_area, and common_mistakes fields defined in SKILL.md.
Boundary markers: Absent; user-provided data is interpolated directly into the markdown prompt structure without explicit delimiters or escaping instructions.
Capability inventory: None; the skill has no tool-calling capabilities, file system write access, or network execution power.
Sanitization: Absent; no validation or filtering is performed on input variables.
Analysis: Due to the lack of capabilities (no code execution or network access), the risk of indirect injection is negligible and is considered safe.

source-credibility-evaluation-protocol