The Agent Skills Directory

[PROMPT_INJECTION]: The skill uses instructional language such as "Hard rules" and "Always pin the API version" to define development best practices for Zotero. No attempts to override agent safety guidelines or bypass instructions were found.
[CREDENTIALS_UNSAFE]: The documentation contains examples using placeholders like <KEY>, <userID>, and <groupID>. No hardcoded API keys or sensitive secrets were detected.
[EXTERNAL_DOWNLOADS]: The skill references official and well-known resources, including Zotero's official GitHub repositories (zotero/zotero, zotero/translators, zotero/make-it-red) and the Citation Style Language repositories. These are treated as trusted sources for development documentation.
[DATA_EXFILTRATION]: The skill provides examples of interacting with the Zotero Web API (api.zotero.org) using curl. These are standard API interaction patterns for reference manager integration and do not involve unauthorized data movement.
[OBFUSCATION]: All files were scanned for Base64 encoding, zero-width characters, homoglyphs, and other obfuscation techniques. No hidden content or deceptive formatting was found.
[DYNAMIC_EXECUTION]: The documentation mentions Zotero's internal "Run JavaScript" tool and API methods like saveTx(). These are legitimate platform features for Zotero developers and are presented for educational purposes only.
[INDIRECT_PROMPT_INJECTION]: While the skill documents tools that ingest external data (Zotero Web API responses and web scraping via translators), it functions as static reference documentation. It does not provide the agent with instructions to process untrusted data without boundaries.

zotero-dev-rules