Product Docs Weekly — issue 004, July 29, 2026

Adds qq_id identifier type to look up QQ/Qzone accounts by numeric ID, extracting nickname and avatar via the Qzone portrait API.

Adds bilibili_id identifier type to look up Bilibili users by numeric mid via the Bilibili user-card API, extracting nickname, avatar, signature, and follower count.

Adds identifier-type filtering so searches automatically exclude sites whose ID format doesn't match the supplied identifier; supported types include gaia_id, steam_id, vk_id, yandex_public_id, ok_id, wikimapia_uid, uidme_uguid, yelp_userid, orcid, qq_id, and bilibili_id.

New Jira Cloud app (built on Atlassian Forge) automatically synchronizes BloodHound Enterprise Attack Path findings to Jira Software and Jira Service Management issues.

Supports AWS STSSTSAWS Security Token Service, an Amazon Web Services API that issues short-lived credentials for IAM roles, letting cyber tools authenticate to AWS without storing long-term secrets. session tags for Bedrock: metadata sent via x-portkey-metadata is forwarded as STS session tags, enabling spend segregation in AWS Cost Explorer and CloudTrail without separate IAMIAMIdentity and Access Management — a framework of policies and services for controlling who can authenticate to systems and what resources they are authorized to use, letting cyber tools enforce least-privilege and audit access. roles per application.

Clarifies that model conditions and group_by keys use the request model name (e.g. gpt-4.1), not the provider's response model name (e.g. gpt-4.1-2025-06-01), which matters for Azure OpenAI and similar providers.

Adds scan_scope and strip_scaffolding parameters to the Palo Alto Networks Prisma AIRS guardrail, enabling fine-grained control over which messages are scanned and reducing false positives for agentic clients like Claude Code, Cursor, and Cline.

Adds soft_deny_200 flag on guardrail checks: denials return HTTP 200 with the error as a chat completion response instead of HTTP 446, keeping sessions alive for clients that treat 4xx as fatal.

GET /v1/models now routes through the gateway's provider-backed model listing handler, calling the upstream provider and returning a normalized model list with response transforms applied.

Anthropic cost attribution now differentiates batch (50% discount), fast, and standard pricing tiers for multidimensional cost tracking.

New integration guide for Continue.dev: adds enterprise-grade observability, cost tracking, and governance to the open-source AI coding assistant via Portkey as an OpenAI-compatible endpoint.

New Admin API endpoints to manage SCIMSCIMA standard for creating, updating and removing user accounts and groups across applications. It helps an identity provider keep access in sync when people join, change roles or leave. group-to-workspace mappings (create, list, delete) for enterprise identity provisioning.

New Admin API endpoints to create and list Rate Limits Policies and Usage Limits Policies, with entity tracking for usage limits.

New Admin API endpoints for custom model management: delete custom models, list model access, and update model access and pricing configuration.

Supports publishing traditional/tabular ML model predictions to Openlayer for monitoring, streamed in real-time or uploaded in batches.

Introduces Snyk Studio, an agentic security platform with quickstart guides for 20+ AI coding agents including Cursor, Devin, GitHub Copilot, Claude Code, Gemini CLI, and Amazon Q.

Adds Evo by Snyk, a new AI Security Posture Management (AI-SPM) platform surface with inventory, policies & issues, reports, AI asset visibility, and risk intelligence.

Launches Agentic Development Security (ADSADSAlternate Data Streams: an NTFS feature that lets a file carry hidden secondary data blobs under the same path, giving cyber tools a place to stash metadata or detect covert storage by malware.) within Evo, covering Agent Supply Chain Security, Agent Behavior Governance, and Trusted Output Assurance.

Adds support for scanning API targets using Bruno Collections in OpenCollection format, including dynamic token generation for authentication.

Introduces a Unified IDE Configuration Dialog for setting authentication, CLI, trust, and Project defaults across Eclipse, JetBrains, Visual Studio, and VS Code Snyk plugins.

Surfaces Non-Human Identity policy breaches (overprivilege, stale identities, cross-environment usage) as actionable incidents with lifecycle tracking.

Replaces the Slack webhook notifier with a native Slack App that threads incident events under a single parent message per incident.

Adds interactive Slack Block Kit buttons (Resolve / Ignore) so developers can act on incidents directly from the Slack notification card.

Delivers incident notifications as Microsoft Teams direct messages to the responsible developer, with configurable Channel / DM / Both delivery mode and a fallback user.

Adds a GitGuardian Power for Amazon Kiro with secret-detection skill guidelines, MCP configuration, and hooks for AI-assisted development.

Adds new Honeytoken deception primitives: fake service endpoints, database credentials, and SaaS API keys.

Enables planting Honeytokens inside HashiCorp Vault, AWS Secrets Manager, and Azure Key Vault as a second dissemination surface.

Adds a unified protection coverage dashboard across all asset types: endpoints, vaults, CI/CD, cloud accounts, and repos.

Introduces an Attack Detection incident type that surfaces Honeytoken triggers as high-priority alerts with attacker forensics.

Adds Public Monitoring to self-hosted instances (license-gated): monitor public perimeter for secret leaks and route incidents through custom webhooks, Slack, Microsoft Teams, ServiceNow ticketing, and severity-filtered email alerts via the new celeryWorkers.public-monitoring worker.

Adds native Gerrit VCSVCSVersion control system: software that tracks changes to files over time, letting teams collaborate and roll back mistakes. Cyber tools use it to manage source code, configs, and signatures. source with full historical scanning, real-time detection via webhook plugin, and read-replica support for cloning.

Adds Smart Notifiers: define per-channel filtering rules so alerts, tickets, or emails are only created for incidents matching severity, validity, secret type, or GitGuardian tag — across internal and public monitoring.

GitHub check runs now scan oversized pull requests by cloning the repository branch instead of calling the GitHub API commit-by-commit, and can skip merge commits; thresholds and rate-limit behavior are configurable from the admin area GitHub preferences.

Adds Gateway API routing: route external traffic through the Kubernetes Gateway API instead of the legacy Ingress resource, with HPA or KEDA autoscaling for webapps.

Adds 9 new detectors (Amadeus OAuth Credentials, Kinde OAuth Credentials, TomTom API Key, Datadog Personal Access Token, Azure Maps, Azure Mixed Reality, Azure Service Bus, Azure Quantum, Azure Bot Direct Line) and 2 new checkers (Gemfury Deploy/Push Token, Gemfury Full Access Token).

Jira Notifier now exposes remediation guidance and detector-specific links as assignable fields in the ticket description template.

Daily analytics job can now write Spark scratch data to a dedicated ephemeral volume, preventing failures from node ephemeral-storage limits on large workspaces.

Fine-grained GitHub personal access tokenspersonal access tokensA personal access token, a credential used by scripts and integrations instead of a user's password. Its permissions and expiry depend on the service issuing it. can now be used to monitor private members of a GitHub organization (previously only classic tokens were supported) in the Public Perimeter view.

Perimeter table now supports column customization, letting users choose which source attributes are displayed to tailor the view to their workflow.

Bulk actions on perimeter sources now support Monitor, Unmonitor, and Assign Team operations on selected sources, with a confirmation popup when selecting more than 30 sources.

Historical scans now have their own dedicated tab on the perimeter page for tracking past and in-progress scans and their statuses.

Adds @mention support in incident notes and feedback — type @name or @team-name to notify individual members or entire teams via email, linked directly to the incident.

Adds includeMcpTools option to @cloudflare/think to opt out of Think's automatic MCP tool exposure when tools are managed externally (e.g., via Code Mode).

Agents SDK MCP clients now reuse converted input/output schemas across model turns within a live connection, eliminating redundant JSON Schema-to-Zod conversions.

Adds execute(), search(), and describe() methods to the @cloudflare/codemode durable runtime handle, enabling MCP servers and non-AI-SDK hosts to invoke Code Mode directly.

Introduces Devin Outposts: run Devin agent sessions on Cloudflare-managed, container-isolated infrastructure via the Sandbox SDK.

Deploys new WAF rules covering Next.js DoS (CVE-2026-64641), Next.js SSRFSSRFServer-side request forgery, an attack that tricks a server into making requests chosen by an attacker. Those requests can reach internal services the attacker cannot access directly. in Server Actions (CVE-2026-64649), Next.js information disclosure (CVE-2026-64643), Adobe ColdFusion path traversal (CVE-2026-48276, CVE-2026-48282), and WordPress unauthenticated SQL injectionSQL injectionAn attack that makes an application treat untrusted input as part of a database command. It can expose or modify data by changing what the query does. (CVE-2026-60137).

Introduces the Permission Groups API as the replacement for the deprecated Account Roles API, supporting API Token authorization and a new meta object with label and scopes in the response schema.

Documents minimum node versions for API Discovery across all supported protocols: GraphQL (NGINX Node 6.1.0 / Native Node 0.15.1), SOAP (NGINX Node 6.3.0 / Native Node 0.17.1), and gRPC (NGINX Node 6.4.0, NGINX only).

Adds structured fuzzy match results and a diagnostics API: every fuzzy match now produces a structured record (rule, symbol, upstream, digests, match type, probability, score, flag, timestamp) accessible via task:get_fuzzy_results() in Lua; a new fuzzy_check.check() Lua API queries a storage with task-generated or explicit hex digests.

Adds rspamadm fuzzy_hash command to compute a message's fuzzy hashes per configured rule, optionally check them against storage (-C), and query storages for explicit hex digests (-H).

Adds rspamadm control fuzzyhash <128-hex-digest> per-hash storage introspection across all fuzzy workers, reporting flag slots, values, creation time, TTLTTLTime to live, a limit on how long data remains valid or how far a packet can travel. Caches use a time limit; IP packets use a hop limit., and shingle slot ownership.

Adds configurable fuzzy probability weight curve via prob_bias and prob_power per-rule settings, replacing the fixed sqrt(prob) multiplier with a threshold-anchored curve.

Adds persistent shingle sets in the Redis fuzzy backend (S field on every ADD), enabling correct digest-only deletions and shingle-anchor refresh.

Adds unkeyed client statistics: traffic from clients without a key is now tracked under a unkeyed pseudo-key with per-IP and aggregate counters.

Adds ClickHouse fuzzy match export: new Fuzzy.Hash, Fuzzy.Queried, Fuzzy.Rule, Fuzzy.Prob, and Fuzzy.Flag nested columns (schema version 11) populated from structured fuzzy results.

Adds annotated X-Rspamd-Fuzzy milter header built from structured fuzzy results, including rule, flag, probability, queried hash, and storage timestamp per match.

Adds forbid_local option to rspamd_http.request() to block SSRF via loopback, link-local, and RFC1918RFC1918IETF standard that reserves three IPv4 address ranges (10.x, 172.16–31.x, 192.168.x) for private networks, never routed on the public internet, giving tools a reliable way to identify internal traffic. destinations; enabled by default in url_redirector.

Adds ARF report enrichment in lua_feedback_parsers: recovers Source-IP, Arrival-Date, Original-Mail-From, and Original-Rcpt-To from the embedded original message headers for Microsoft JMRP feedback loops.

Adds composite Lua conditions: rspamd_config:add_composite() now accepts a conditions map of synchronous Lua functions gating atom evaluation, with optional depends_on for cross-pass symbol access.

Introduces policy = "implicit_allow" in settings apply blocks, allowing symbols_enabled to unlock explicit_enable symbols without switching the task into whitelist mode.

Adds static_embed neural provider: a Model2Vec-style, mmap-shared float32 token-embeddingembeddingLists of numbers a model produces to represent the meaning of text, images or code, so similar items sit close together. They power semantic search and RAG. backend requiring no neural forward pass, no FFI, and no new dependencies, with a get_token_vectors() method for external consumers.

Adds train.forced_learn_minimal_scan neural training option that installs a prefilter disabling all non-neural symbols during ANN-Train scans, skipping RBLRBLA DNS-based blocklist of IP addresses known to send spam or host malicious traffic, maintained by various operators, used by cyber tools to reject or flag connections from listed sources at lookup speed./DNS/fuzzy/Bayes work.

Adds train.frozen neural training option that halts automatic training and vector storage while keeping inference active; explicit ANN-Train requests still retrain on demand.

Enables external rspamadm sub-command discovery from a rspamadm.d drop-in directory and the RSPAMADM_COMMAND_PATH environment variable, allowing third-party packages to ship commands without touching rspamd's lualib tree.

Supports supplying the controller password via environment variable for non-interactive automation of password-protected rspamadm operations.

Adds --symbol-details flag to rspamadm configdump, reporting a flags array per symbol (fine, explicit_enable, composite, coro, etc.).

Migrates all WebUI tables from FooTable to Tabulator, adding per-column header filters, a global boolean search box (AND/OR/-/quoted phrases) for history and scan views, persisted column visibility, and automatic light/dark theming via Bootstrap CSS variables.

Exposes aliased and original email-address flags to Lua and adds the orig address flavour to selectors as from('mime', 'orig').

Invalid selectors now fail configuration load with an error pointing at the offending token, preventing silently truncated selector chains in multimap, rbl, ratelimit, settings, reputation, and the WebUI.

Glob map patterns are now compiled anchored (^(?:...)$) at load time, closing a subdomain evasion vector where bare entries like *.bit.ly previously matched foo.bit.ly.evil.com.

Symbols carrying the explicit_enable flag are now strictly gated: they run only when listed in symbols_enabled of applied settings, with a new policy = "implicit_allow" option to unlock symbols additively without whitelisting.

lua_cryptobox secretbox now correctly zero-pads short nonces to 24 bytes, producing stable, reproducible ciphertexts instead of memory-dependent output.

Adds configurable support message (up to 500 characters) displayed to members when a sandbox action is blocked by organization governance policy, enabling admins to direct users to internal support channels or security contacts.

Supports connecting the ChatGPT desktop app to a Docker Sandbox over SSH so OpenAI Codex executes inside the isolated environment rather than on the host machine.

Enables Claude Desktop to run Claude Code inside a Docker Sandbox over SSH, keeping agent activity isolated from the host machine.

Adds SSH connection workflow in Claude Desktop: select a named sandbox as the remote environment via the environment drop-down and SSH Host field.

Provides Windows-specific troubleshooting guidance for SSH connection timeouts caused by missing sh.exe (Git for Windows dependency).

Connects the Cursor editor to a Docker Sandbox over Remote-SSH, keeping the editor on the host while files, terminals, and extensions run inside the isolated sandbox.

Supports connecting VS Code to a running sandbox via the Remote-SSH extension, so files, terminals, and extensions run inside the isolated environment while the editor stays on the host.

New sbx setup ssh command configures SSH access to sandboxes, writing a managed block to ~/.ssh/config so any SSH-capable editor or app can connect.

Sandboxes are now reachable as SSH hosts via <name>.sbx naming, enabling remote development from VS Code, Cursor, Claude Desktop, ChatGPT, and any other SSH-over-remote tool.

SSH connections route through the daemon's local socket (Unix domain socket on macOS/Linux, named pipe on Windows) — no open network port or stored SSH key required.

Environment variable forwarding via SendEnv * is supported, filtered by a daemon-side ssh.acceptEnv allowlist; sensitive variables like PATH, LD_*, and NODE_OPTIONS are always blocked.

Adds sbx skills import command to copy agent skills (Claude Code, Codex, Copilot, Cursor, Droid) from host into a persistent shared store mounted into new sandboxes automatically.

Adds --no-share-skills flag to sbx run to create a sandbox that opts out of the shared skills store, keeping it outside the shared trust boundary.

Supports zero-offset timezone names (GMT, UTC, UT) in timestamp parsing, enabling out-of-the-box handling of web server and mail log timestamps.

Adds automatic type detection for GMT/UTC/UT-suffixed values (e.g., in CSV files), promoting them from strings to timestamps.

The time() function now emits a warning with the offending value when a string cannot be parsed, instead of silently returning null.

New from_splunk operator runs bounded searches against a Splunk Search Head and emits results as events, with support for time-bounded queries using native Splunk relative-time syntax, secret-valued authorization headers, TLS, timeouts, and retries.

The to_clickhouse operator now internally re-batches events before inserting, accumulating events per target table and flushing at max_batch_rows (default 65536) or batch_timeout (default 1s) for more efficient inserts with heterogeneous input like OCSFOCSFAn open schema standard, maintained by the OCSF community under CISA and industry backing, that defines a common event format so cyber tools can share and ingest log data without per-source normalization work..

The to_clickhouse operator can now write string values directly into a ClickHouse JSON column when the string is a valid JSON object, enabling single-schema insert workflows for heterogeneous events.

New MCP server lets external AI clients (Cursor, Claude Code, Codex) connect to Cribl environments and interact with Sources, Destinations, Pipelines, and Search Notebooks via natural language.

Guard recommendations surface a suggested action (create a rule, mark mitigated, or ignore) for each detection so practitioners can act on findings without leaving the UI.

Bring Your Own Model (BYOM) support lets operators route Cribl AI through their own cloud-hosted or self-hosted AI provider instead of Cribl-managed models.

AI feature entry points remain visible and surface a actionable error message when a configured custom provider is unreachable, rather than failing silently.

Adds Chrome Enterprise Premium integration with curated detections enriched by Safe Browsing and Google Threat Intelligence (GTI) to surface browser-based threats in Google SecOps.

Adds dedicated Google SecOps dashboards for Chrome security events and anomalies to improve analyst visibility into browser activity.

Enables automated SOARSOARSecurity orchestration, automation and response, coordinating incident workflows across tools. Playbooks can gather evidence, assign work or carry out approved response actions. playbook response actions on Chrome Enterprise — including blocking malicious URLs, blocking extensions, and quarantining devices — directly from Google SecOps.

Adds Cloud Detection, Investigation, and Response (CDIR) capability spanning Google Cloud, AWS, and Azure, with Security Command Center findings automatically mapped to Google SecOps cases.

Adds Cloud Asset Inventory collection that flows multi-cloud asset metadata (AWS, Azure, Google Cloud) into the Google SecOps Entity Graph for enriched alert context.

Supports natural language query syntax alongside existing Yara-L and raw log search for SIEM investigation during CDIR workflows.

Integrates AI Protection into Security Command Center Enterprise (SCC-E), providing discovery of AI asset inventory, virtual red teaming to surface toxic attack pathsattack pathsA possible sequence of steps an attacker could take to reach a valuable target. It connects weaknesses, permissions and reachable systems rather than treating each finding separately., and detection and response for AI-targeted threats (LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. hijacking, data poisoning, exfiltration).

Adds Model Armor for model- and cloud-agnostic protection against data loss, brand damage, and offensive content in generative AI workloads.

Adds Advanced Filtering in dashboards (Public Preview): security analysts can define Token Variables to inject dynamic values, complex regular expressions, or boolean logic directly into YARA-L queries at runtime.

Customizable prefix/suffix wrappers on token values enable advanced patterns such as inline regular expressions within dashboard queries.

Multi-select support for token variables allows combining multiple values in a query using a configurable delimiter (e.g., |).

Introduces Security Tokens for metering agentic consumption within Google SecOps, consumed by generally available security agents.

Supports the OWASP Top 10 for Agentic Applications framework, enabling practitioners to view, assess, and improve security configurations against agentic AI risks.

Adds initial Apple Metal graphics framework support via software-based implementation, enabling Flutter-based iOS apps that previously failed to launch or render in virtualized environments.

Introduces Model Context ProtocolModel Context ProtocolModel Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. (MCP) Server integration, allowing AI-powered tools and autonomous agents to interact directly with Corellium environments through a standardized interface.

Adds Android Kernel Address Sanitizer (KASAN) and Hardware-Assisted Address Sanitizer (HWASAN) kernel debugging support for identifying memory corruption, use-after-free, and buffer overflow issues.

Enables pre-booting supported Apple virtual devices directly into PongoOS before loading iOS, streamlining jailbreaking and low-level security research workflows.

Supports downloading entire folders and nested sub-folder structures from virtual devices via the Files Browser, accelerating forensic evidence collection.

Adds AI Summary panel on asset detail pages that provides a concise overview of identity, business context, operational context, associated users, configured mitigations, and detected vulnerabilities.

AI summaries are generated on first open and cached for faster subsequent access, with automatic regeneration when new data is ingested.

Adds thumbs-up/thumbs-down feedback form on AI summaries, allowing practitioners to rate and annotate summary quality — submittable multiple times per asset.

New hosted Adversary Investigation MCP server at https://mcp.platform.censys.io/threat-hunting/mcp/ enables AI coding agents to query Censys threat-hunting data.

Supports OAuth and HTTP header authentication for integrating the MCP server with tools like Cursor, Claude Code, and Continue.dev.

New Connected Apps page in Account Settings lets users audit and revoke OAuth 2 applications — including the Platform MCP server — that have been granted access to Censys APIs on their behalf.

Adds a user-facing chat agent with three tools: natural-language BQL query generation, documentation Q&A, and CVE exploration.

Adds an Attribution Agent that autonomously infers Remediation Owner, Business Owner, and Application Owner attributes and contributes them to the attribute consolidation hierarchy.

New MCP server integration exposes BrinqaIQ capabilities — BQL generation, BQL execution, CVE exploration, docs search, and automation lifecycle management — to any MCP-compatible AI client.

Supports natural-language BQL query generation and direct BQL execution with paginated results for feeding Brinqa data into external agent pipelines.

Automation lifecycle control (launch, resume, cancel, status check) and flows/automations discovery (list, inspect by name or ID, list running flows) available through MCP-connected clients.

CVE Explorer tool enables connected AI clients to look up CVE details, severity, affected products, and remediation context.

Supports Claude Desktop, GitHub Copilot in VS Code, Goose, and Open WebUI as tested MCP clients; Open WebUI connects via OpenAPIOpenAPIA vendor-neutral specification for describing REST APIs in a machine-readable format, maintained by the OpenAPI Initiative. Cyber tools use it to auto-generate clients, documentation, and validation from a single source of truth. spec rather than stdio transport.

API token-based authentication ensures all MCP calls inherit the user's existing RBACRBACRole-based access control, where permissions are assigned to roles such as reader or administrator and users receive those roles. It makes access easier to manage than individual permission lists. and data-level permissions from the Brinqa Platform.

New BQL API (POST /v1/api/bql) replaces the legacy GraphQL API, providing programmatic access to the Brinqa data warehouse using the same query engine as the UI.

Supports asynchronous two-step query workflow: submit a query and receive a polling Location URL, then GET results with automatic Retry-After guidance.

Introduces two query modes — Analytics mode (full BQL with RETURN/ORDER BY/LIMIT in-query, returning flattened rows) and Projection mode (returningFields/orderBy/limit parameters, returning nested relationship objects).

Projection mode supports nested relationship projections and wildcards, enabling structured retrieval of one-to-many relationships as arrays within a single result row.

Cursor-based pagination allows retrieval of large result sets up to 10,000 rows per page, with cursor and Location/Link headers pointing to subsequent pages.

New API token authentication system (ApiKey scheme with brq_-prefixed tokens) tied to user ACLACLAn access control list, a set of rules allowing or denying operations for particular users, services or network addresses. It decides who can use a resource./RBAC permissions, with configurable expiration (1/3/6/12 months or never) and self-service revoke/reinstate/delete lifecycle management.

New IONIX connector ingests Action Item, Action Item Definition, Cloud Asset, Domain, IP Address, Managed Domain, and Sub Domain data into the Brinqa graph.

Adds KYB with Risk Assessment integration, enabling business entity risk reviews alongside identity verification workflows.

New MCPMCPModel Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. server at https://documentation.idenfy.com/mcp connects AI assistants (Claude, Cursor, ChatGPT, Gemini, Perplexity, Zed, Windsurf, VS Code) to the full iDenfy documentation.

Exposes search_idenfy_documentation tool: searches all guides and API references, returning page titles, paths, and short snippets.

Exposes get_page_idenfy_documentation tool: retrieves full page content by path (e.g. kyc/generate-token) for deep reference lookup.

Covers all iDenfy products: Identity Verification (KYC), Business Verification (KYB), AML Screening and Monitoring, Fraud Prevention, Face Authentication, Bank Verification, Mobile SDKs, and API Reference.

Enables AI-assisted KYC/KYB integration code generation, webhook debugging, error-code lookup, and compliance Q&A grounded in current documentation rather than stale training data.

New Bank Card Verification step confirms the card used during onboarding belongs to the verified user, configurable as an additional stage in the KYC workflow.

New Suspected Status reference documents why a verification resolves to SUSPECTED and which tags trigger it, aiding triage of edge-case outcomes.

New Status Handling guide maps every iDenfy verification status to recommended actions, covering webhook, API, and dashboard review paths.

New Supported Browsers reference documents which browsers are compatible with iDenfy's redirect and iFrame identity verification integrations.

Supports Zapier Webhooks integration, enabling identity verification events to trigger automated workflows across thousands of connected apps.

New POST /v3/phone/verify/ endpoint verifies OTPs and returns phone intelligence: carrier name, line type, VoIP/disposable flags, and duplicate usage across sessions.

New POST /v3/email/verify/ endpoint verifies OTPs and returns email intelligence: deliverability status, breach exposure (is_breached + breaches list), disposable-provider detection, and duplicate usage across sessions.

Configurable risk actions (duplicated_phone_number_action, disposable_number_action, voip_number_action for phone; duplicated_email_action, breached_email_action, disposable_email_action for email) let you set DECLINE or NO_ACTION per risk type to control whether a detected risk terminates or merely flags a verification.

3-attempt budget enforced on both phone and email OTP checks: first two wrong codes return Failed with remaining attempts; third wrong code finalizes as Declined with a dedicated warning code.

Sandbox mode for both verify endpoints: any well-formed code returns a static Approved payload without persisting state or calling providers.

Finalized OTP checks persist to the originating session (visible in Business Console, queryable via GET /v3/session/{sessionId}/decision/) and fire a status.updated webhook.

Backend-only resubmit throttle: resubmitting a session for backend-only features (AMLAMLAnti-Money Laundering: a set of laws, regulations, and procedures that financial institutions must follow to detect and report suspicious financial activity; cyber tools use it to automate transaction monitoring and compliance workflows., DATABASE_VALIDATION, IP_ANALYSIS) within 30 seconds now returns a structured error with a Retry-After interval instead of silently repeating.

Adds a pre-registration discovery check that surfaces relaxable authorization server issues (e.g., issuer mismatch, cross-host auth server) for admin acknowledgment before proceeding, and blocks registration entirely for issues that cannot be safely relaxed.

Supports an optional tool prefix override on MCP server registration; falls back to the server's declared prefix, then a C1-generated default.

Adds per-server tool enablement setting (Require enablement or Auto-enable) that overrides the tenant-wide default tool approval setting for a specific MCP server.

Supports side-by-side multi-chart dashboards in a single query (e.g., trend chart alongside a users-by-status chart), each with its own export option.

Adds a View lineage Data lineage drawer showing the exact query and row count behind each report source, verification status against the assistant's execution log, and the full list of tool calls made to build the report.

Enables copying individual source queries or the entire lineage (sources, verification status, tool calls) as Markdown for audit workpapers.

Assigns a workflow step in a request or review policy to C1 AI, which evaluates each request using built-in guidance plus custom instructions; every evaluation is logged as an AI-agent action in the audit log and recorded as a reviewable conversation.

Adds an indirect-access section to entitlement search results, surfacing entitlements (e.g., a GitHub team) that transitively grant access to a searched resource.

Enables custom recurrence intervals on campaign templates (e.g., every two days, every three months) and lets admins set exactly how long each campaign run lasts, replacing fixed presets.

Adds a button to open raw access-review and task-detail data as a readable, collapsible JSON tree on resource-level access reviews and in task details.

Supports adding groups to a request catalog or access profile so users can request group membership through the same approval flow as other requestable access.

Holds provisioning requests for entitlements not yet recognized by C1 (e.g., created via Terraform before a connector sync), shows the reason for the wait, resumes automatically after the next sync, and falls back to manual provisioning after 24 hours.

Supports combined step-up (MFAMFAMulti-factor authentication, requiring proof from different categories, such as a password and a hardware key, to sign in. It reduces reliance on a password alone.) authentication and comment requirements on a single approval task, prompting for identity-provider verification and completing once verified.

Adds a View lineage action on C1 AI-generated reports to inspect the exact queries behind each number, verify each source against the execution log, and copy lineage for audit workpapers.

Allows admins to require per-user connector credentials (OAuth2OAuth2An open authorization framework (RFC 6749) that lets a service grant limited access to a user's resources on another service without exposing credentials, giving cyber tools a standard way to authenticate against third-party APIs., bearer tokenbearer tokenA credential that grants access to whoever presents it, without a separate proof of possession. Protecting the token matters because a stolen copy can be used by someone else., custom header, or basic auth) instead of a shared admin credential.

Adds automated scheduling for campaign templates, enabling draft campaign instances to be created once on a future date or on a recurring schedule.

Supports configurable recurrence intervals (e.g., every 2 days, every 3 months) and campaign duration settings via the template's Automated scheduling section.

New read-only AI > Conversations page lets admins review AI agent conversations across the organization, organized into separate tabs by surface.

New System tab surfaces agent policy step evaluations — including the agent's reasoning and tool calls — attributed to the built-in C1 System identity for every approval policy decision.

Adds Retry failed provisioning action to re-attempt a failed connector provisioning step on an access-change task without restarting the approval workflow.

Adds a View policy button in the resource details drawer that opens attached policy documents (e.g. AWS IAMAWS IAMAWS Identity and Access Management is Amazon's service for controlling who and what can call AWS APIs, by attaching policies that allow or deny actions on resources. Cyber tools use it to scope credentials to least privilege. policies) as a readable, collapsible JSON tree instead of raw text.

Assign an AI Agent step to any request or review policy workflow, configuring agent mode, failure action, and free-text approval/deny instructions; every evaluation is attributed to the C1 System identity and logged under AI > Conversations.

Catalog search now returns an Entitlements matching section showing entitlements that indirectly grant access to the searched resource, with a View access graph link to see the full grant chain.

Access review campaigns can now review grants held by groups or roles (not just users) via a new Grant Types selector offering Users only, Resources only, or Users and resources.

Campaign templates support custom recurrence intervals (e.g., every 2 days or every 3 months) and configurable campaign duration, beyond the previous fixed presets.

IAMIAMIdentity and Access Management — a framework of policies and services for controlling who can authenticate to systems and what resources they are authorized to use, letting cyber tools enforce least-privilege and audit access. policy documents on AWS resources now render as a readable, collapsible JSON tree via a View policy button, available in resource-level access reviews and task details.

Requests for entitlements created ahead of a connector sync (e.g., via Terraform) now wait and resume automatically after sync completes, with a 24-hour manual-provisioning fallback.

C1 groups can now be added to a request catalog or access profile, making group membership requestable through the standard approval, duration, and revocation flow.

Failed connector provisioning on an access task can now be retried directly via Retry failed provisioning without cancelling the task or losing collected approvals.

External MCP servers now support per-user authentication (OAuth2, bearer token, custom header, or basic auth) instead of requiring a shared admin credential.

AI-generated reports include a Data lineage view showing the exact queries behind each number, verified against the execution log, with copy support for audit workpapers.

Ask C1 AI now renders multiple requested charts as an aligned grid and can produce composed reports combining stat cards, a trend chart, and a data table in one response.

The CELCELCommon Expression Language, an open-source expression language maintained by Google, used to embed lightweight, safe policy evaluation logic into tools without running a full scripting engine. editor gains a Generate action that converts a plain-English description directly into a CEL expression in one step.

Tool enablement (require enablement vs. auto-enable) can now be set per MCP server, independent of the tenant-level AI governance default.

External and hosted MCP servers that fail relaxable OAuth discovery checks (e.g., issuer mismatch) can now be registered after an admin reviews and accepts the flagged findings.

Hosted MCP servers now default to using the tool prefix declared by the connector itself rather than a generic derived name.

Any user can now use the Limit Scopes selector when creating a personal API key to restrict it to a narrower role (e.g., read-only), without requiring Super Admin access.

Adds a Grant type attribute to campaigns and campaign templates, enabling access reviews scoped to users only, resources only (groups/roles), or both users and resources.

Documents the C1 MCP integration as a programmatic path for external AI clients (such as Claude Desktop or Cursor) to query C1 identity data.