Setup AI

Apex Recon integrates with two AI providers — pick whichever account you have credit on:

• OpenAI (openai.com) — openai provider.
• Anthropic Claude (anthropic.com) — claude provider.

Configuring a provider unlocks Djin! — the in-app side-dock chat assistant — across the Pro web UI. Apex’s domain is input-vector discovery (not security findings to triage as resolved), so there’s no per-record analysis plugin the way Spectre Scan ships one; configuring the AI key here enables only Djin!.

Web UI

In Settings → Djin! AI assistant:

1. Pick a provider (OpenAI or Claude) from the dropdown.
2. Paste your API key. Apex Recon Pro pings the provider on save to verify the key — invalid keys / billing issues / no model access surface as a precise error rather than a silent failure.
3. Save.

That alone enables Djin!. Two extra controls in the same section tune behaviour:

• Djin! assistant — toggle. Off by default; flip on once the provider key validates. Off renders the dock as an empty “Djin! is disabled — enable in settings” hint instead of failing silently.
• Djin! daily token budget — per-user 24-hour cap on prompt
  • completion tokens summed across every Djin! conversation (default 500 000). 0 disables the cap entirely.

What Djin! does

See the dedicated section in the Pro features page for the full surface — page-context grounding, the read tools (list / fetch / search / events / sitemap / dashboard / …), the write tools (mark_reviewed, scan lifecycle, quick_scan, create_site_role, …) gated behind explicit Approve clicks, and the PaperTrail audit-trail tagging that distinguishes AI-driven mutations as whodunnit = "djin:<your_id>".