Ai Puffer Ai Chatbot Ai Writer Automation
Monthly
Authenticated users with subscriber-level access or higher can alter site-wide semantic search configuration in the WordPress plugin AI Puffer (formerly AI Power) when an administrator has already granted their role the Knowledge Base ('sources') module through the plugin's Role Manager. The flaw affects all versions up to and including 2.4.89 and allows an attacker to change global values in the aipkit_options record - vector provider, embedding provider, embedding model, target ID, result count, and no-results text - that are otherwise administrator-only, giving a limited integrity impact with no confidentiality or availability loss. Exploitation requires a non-default role grant, so reachability is constrained; there is no public exploit and no confirmed active exploitation (no CISA KEV listing) at time of analysis, and the CVSS 3.1 vector scores this as low severity (3.1, AV:N/AC:H/PR:L/UI:N/S:U/C:N/I:L/A:N).
Authorization bypass in the WordPress plugin AI Puffer - Chat. Create. Automate. (formerly AI Power, slug gpt3-ai-content-generator) through version 2.4.89 lets an authenticated subscriber-level user rewrite the global options aipkit_training_general_settings and aipkit_indexing_field_settings, changing chunking parameters, file upload visibility and per-post-type field indexing for the whole site. Exploitation is tightly constrained: the attacker must already hold an account (PR:L per CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:N/I:L/A:N) and an administrator must have explicitly granted the 'sources' module to that lower-privileged role through the plugin's Role Manager, a non-default configuration against which the plugin is not exploitable at all, which is why attack complexity is rated High. Impact is confined to configuration integrity with no data disclosure or availability loss, no public exploit identified at time of analysis, and this is assessed as a genuinely low-priority issue.
Authenticated users with subscriber-level access or higher can alter site-wide semantic search configuration in the WordPress plugin AI Puffer (formerly AI Power) when an administrator has already granted their role the Knowledge Base ('sources') module through the plugin's Role Manager. The flaw affects all versions up to and including 2.4.89 and allows an attacker to change global values in the aipkit_options record - vector provider, embedding provider, embedding model, target ID, result count, and no-results text - that are otherwise administrator-only, giving a limited integrity impact with no confidentiality or availability loss. Exploitation requires a non-default role grant, so reachability is constrained; there is no public exploit and no confirmed active exploitation (no CISA KEV listing) at time of analysis, and the CVSS 3.1 vector scores this as low severity (3.1, AV:N/AC:H/PR:L/UI:N/S:U/C:N/I:L/A:N).
Authorization bypass in the WordPress plugin AI Puffer - Chat. Create. Automate. (formerly AI Power, slug gpt3-ai-content-generator) through version 2.4.89 lets an authenticated subscriber-level user rewrite the global options aipkit_training_general_settings and aipkit_indexing_field_settings, changing chunking parameters, file upload visibility and per-post-type field indexing for the whole site. Exploitation is tightly constrained: the attacker must already hold an account (PR:L per CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:N/I:L/A:N) and an administrator must have explicitly granted the 'sources' module to that lower-privileged role through the plugin's Role Manager, a non-default configuration against which the plugin is not exploitable at all, which is why attack complexity is rated High. Impact is confined to configuration integrity with no data disclosure or availability loss, no public exploit identified at time of analysis, and this is assessed as a genuinely low-priority issue.