The Agent Skills Directory

[SAFE]: No security issues detected. The skill serves as a technical guide for Frappe search systems (Link Search, Global Search, Website Search, and SQLiteSearch).- [PROMPT_INJECTION]: No attempts to override agent behavior, bypass safety guidelines, or extract system prompts were found. Additionally, while the skill describes systems that ingest untrusted user input (search queries), it provides explicit security mitigations. 1. Ingestion points: User input via search text in search_link and global search APIs in SKILL.md and references/link-search-api.md. 2. Boundary markers: The skill includes an 'Anti-Patterns' section in SKILL.md explicitly requiring parameterized queries and @frappe.whitelist(). 3. Capability inventory: Database read operations (e.g., frappe.db.sql, frappe.get_all) are used across references/global-website-search.md and references/link-search-api.md; no file-writes or network-sends are present. 4. Sanitization: Documentation in references/global-website-search.md specifies that indexed content is HTML sanitized and emphasizes query parameterization.- [DATA_EXFILTRATION]: No network operations, sensitive file access, or hardcoded credentials were detected. The skill documentation focuses on internal framework APIs.- [COMMAND_EXECUTION]: The skill does not contain shell commands, subprocess invocations, or privilege escalation patterns.

frappe-core-search