The daily firehose — everything the toolchain shipped today, already filtered.
We read every release from the 1309 tools on our watchlist at the source — GitHub and GitLab release notes, vendor release pages and changelogs, project blogs and feeds, vendor press releases, and the source code behind the tag. Bug-fix-only releases and non-product newsroom noise are dropped; what's left is summarized down to the new capability, how to try it, and any screenshots or videos the release itself published. Every entry links to the sources it was built from.
READ AS
Focus:What features shipped and what problems they solve, plus launches, packaging and where the market is moving.
Leaves out:Command-line flags, API-only deltas, usage examples, and releases that are only refinements.
Focus:New security controls, new APIs, and product improvements you can put to work, plus anything an upgrade breaks.
Leaves out:Launch and marketing news, governance, pricing, and documentation changes with no operational effect.
Focus:Dashboards, reporting, and anything you can quantify: metrics, coverage, risk scores, compliance impact, vendor moves.
Leaves out:Settings, flags, API and command-line detail, library bumps, and minor releases with no reporting or posture angle.
Focus:API, SDK and MCP changes, new libraries, breaking changes in what you import, and cyber tooling.
Leaves out:Governance, SOC operations, and vendor launches with no developer surface.
AMOUNT
DETAIL
ORDER
This issue came out before editions were ranked, so it shows the Standard edition in issue order.
Promptfoo is an open-source tool for testing, evaluating, and red-teaming LLM applications to detect vulnerabilities, compare outputs, and ensure AI reliability.
Promptfoo snapshot-20260717 adds adaptive rate limiting, Transformers.js local inference, telecom red team plugins, and video generation providers.
└──▷ TRY IT
Load multiple environment files (e.g., base secrets plus environment-specific overrides) in a single eval run.
Adds Transformers.js provider to run Hugging Face models locally in Node.js or the browser.
Adds Vercel AI Gateway and Cloudflare AI Gateway providers for request routing.
Adds AWS Bedrock Video (Nova Reel, Luma Ray 2) and Azure AI Foundry Video (Sora) for video generation.
Adds native session endpoint support on the HTTP provider for stateful conversations.
Adds xAI Voice provider overrides for apiBaseUrl, websocketUrl, and function call support.
+22 moreshow less
Adds word-count assertion type for validating response word counts.
Adds __count variable for use in derived metrics to compute averages.
Adds provider config hover in eval results UI to view provider configuration details.
Surfaces metadata.sessionId as a variable column in eval tables and exports.
Adds user-rated filter to show only manually rated results in the UI.
Adds promptfoo logs command for viewing log files directly.
Adds adaptive rate limit scheduler that automatically adjusts concurrency based on provider rate limits and response headers.
Adds test-level prompts filter to filter prompts at the individual test case level.
Adds providers filter at the test case level.
Adds per-test structured outputstructured outputForcing a model's response to match a given format or JSON schema, so software can parse it reliably instead of hoping the text is well formed. configuration at the test case level.
Supports $VAR syntax for environment variables in file paths.
Supports multiple --env-file flags to load multiple environment files.
Adds fork PR support and comment-triggered scans for code scanning in CI/CD.
Adds --share and --no-share auto-sharing flags to Model Audit for controlling cloud sharing.
Adds Telecom industry-specific red team plugins for telecommunications AI systems.
Adds RAGRAGRetrieval-augmented generation: looking up relevant documents and adding them to the prompt before the model answers. It grounds responses in private or current data without retraining. Source Attribution plugin to test whether RAG systems properly attribute sources.
Red team scans now support multiple input variables for testing systems with complex input structures.
Adds numTests config to cap test cases generated per red team strategy.
Adds -d/--description flag to redteam generate commands.
Adds early scan stop when plugins fail to generate test cases.
Adds automatic retries for transient 5xx errors.
Propagates -j flag to Python worker pools for improved Python concurrency.
Portkey is an AI gateway and LLMOps platform that helps developers manage, monitor, and optimize calls to large language model APIs with reliability and observability.
Portkey adds vault-backed credentials, weekly rate limits, cross-cloud GCS logging, and four new guardrail integrations in March 2026.
└──▷ HOW TO FIND IT
Pull 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. provider credentials from your existing vault at runtime instead of storing keys in Portkey directly.
📍In the Portkey console, go to Settings › Integrations (or Virtual Keys) and add a secret_mappings entry pointing to your AWS Secrets Manager, Azure Key Vault, or HashiCorp Vault secret ARN/path. Portkey fetches the value at request time.
Route Zscaler AI Guard policy checks across all LLM traffic to enforce your org's existing detection policies on inputs and outputs.
Ship gateway logs from AWS to GCS using Workload Identity Federation — no long-lived GCP keys required.
$ GCP_WIF_AUDIENCE=<your-wif-audience>
GCP_WIF_SERVICE_ACCOUNT_EMAIL=<your-service-account>@<project>.iam.gserviceaccount.com
# Set log storage mode to gcs_assume in your gateway config
New Secret References let you point Portkey at credentials stored in AWS Secrets Manager, Azure Key Vault, or HashiCorp Vault via secret_mappings, so keys are fetched at runtime from vaults you already control.
Adds weekly rate and budget windows (rpw) so spend caps align with weekly planning cycles rather than only per-minute or monthly aggregates.
Adds endpoint-scoped rate limits so chat completions, embeddingsembeddingsLists 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., and admin routes can each carry independent limits instead of a single global rule.
GCP Workload Identity Federation log delivery: configure GCP_WIF_AUDIENCE and GCP_WIF_SERVICE_ACCOUNT_EMAIL so an AWS-hosted gateway writes logs to GCS without long-lived GCP keys (gcs_assume style).
Organization admins can now include archived workspaces in analytics graphs and filters; saved filters also accept workspace slugs alongside IDs.
+9 moreshow less
New Zscaler AI Guard integration applies Zscaler Detection Policies to LLM inputs and outputs via beforeRequestHook / afterRequestHook with a required policyId.
New Akto Agentic Security guardrail scans LLM inputs and outputs for prompt injectionprompt injectionAn attack where text the model reads, such as a web page, file or tool result, carries instructions that override the user's. It is the main security risk for agents with tools. and sensitive data leakage with configurable hook and timeout.
Bedrock Guardrails plugin now accepts a customHost field to route guardrail evaluation to private or regional Bedrock-compatible endpoints.
New required metadata key–value guardrails block requests missing specified metadata fields before they reach the model.
DeepInfra provider adds tool callingtool callingA model's ability to answer with a structured request to run a named function or API, which the application executes and returns. It is how agents take actions. (tools, tool_choice, parallel_tool_calls) plus completions and embeddings endpoints.
DeepSeek deepseek-chat gains tools, tool_choice, and stream_options; deepseek-reasoner maps reasoning_effort to thinking mode and returns reasoning_content in streams.
Bedrock batch inference now supports embeddings in addition to chat completions.
Vertex AI: Portkey metadata maps to Vertex resource labels; enterprise search grounding via enterpriseWebSearch / enterprise_web_search; AWS workloads reach Vertex via AWS–GCP WIFWIFWorkload Identity Federation is a Google Cloud mechanism that lets external workloads authenticate to Google APIs using short-lived tokens from a trusted identity provider, eliminating the need to manage long-lived service account keys..
Azure AI Foundry gains Cohere rerank model support (e.g. cohere.Cohere-rerank-v4.0-pro); gateway strips the cohere. prefix for the provider.
Veracode is a cloud-based application security platform that scans code for vulnerabilities, helping developers identify and fix security flaws across the software development lifecycle.
Veracode adds OAuth for XML APIsXML APIsInterfaces for reading, writing, or querying XML documents, provided by libraries such as libxml2 or built into language runtimes; cyber tools use them to parse structured data like configs, reports, or protocol messages reliably., six new built-in policies, OWASPOWASPA nonprofit foundation that publishes open standards, guides, and ranked vulnerability lists—most famously the OWASP Top 10—giving cyber tools a vendor-neutral benchmark to test or report against./CWECWECommon Weakness Enumeration, a catalog of recurring software and hardware weakness types. It describes classes of mistakes, such as injection, rather than individual vulnerability instances. 2025 standards, and a new Firewall Analytics dashboard.
Supports OAuth Client Credentials authentication for Veracode XML APIs, not just REST APIsREST APIsA convention for building HTTP-based interfaces where resources are addressed by URL and manipulated with standard verbs like GET, POST, and DELETE, giving cyber tools a predictable way to integrate with external services..
Adds six new built-in security policies to the Veracode Platform, ahead of planned deprecation of recommended and transitional policies.
Auto-Update OWASP and Auto-Update CWE Top 25 policy rules now enforce OWASP Top 10:2025OWASP Top 10:2025A ranked list of the ten most critical web application risk categories, maintained by the Open Worldwide Application Security Project, used by cyber tools to prioritize detection, testing, and remediation coverage. and MITREMITREA US federally funded research and development center that maintains public frameworks—ATT&CK, CVE, D3FEND—that cyber tools use to classify threats, vulnerabilities, and defensive techniques in a common language. CWE Top 25 for 2025 standards.
Adds collections_list field to Finding Reporting API responses, returning collection names associated with an application.
Adds Collections fields to the Applications explore in Veracode Analytics, with new Policy Compliance Overview dashboard visualizations.
+18 moreshow less
New most_recent_scan filter for the Reporting API Scans report returns only the most recent scan per asset when set to true.
New Firewall Usage Dashboard and Explore in Veracode Analytics surface metrics for Veracode Firewall.
Policy Settings page adds light and dark mode support.
New library_instance_ref field in Reporting API findings response uniquely identifies the specific library instance associated with a finding.
New custom_severity filter available for the Reporting API Findings report.
New data_refresh_date field added to the Reporting API to indicate the most recent Analytics data refresh time.
Supports OAuth 2.0OAuth 2.0An open authorization framework (IETF RFC 6749) that lets a service grant limited API access to a third party without sharing credentials, giving cyber tools a standard way to authenticate against external platforms. Client Credentials flow for generating and managing API credentials on the API Credentials page in the Veracode Platform.
Veracode Analytics data refresh interval reduced from four hours to two hours.
MPT Flaw Matching correlates Manual Penetration Testing findings across tests for consistent flaw IDs, status values, and remediation grace period inclusion.
Administrators can now disable Veracode-defined roles for users who use custom roles, with corresponding Identity API updates.
Container and IaCIaCInfrastructure as Code: the practice of defining and provisioning compute infrastructure through machine-readable config files rather than manual processes, giving cyber tools repeatable, version-controlled environment setup. scan results from the CLI are now viewable on the Veracode Platform, with severity counts, filtering, scan history, and mitigation guidance.
New action_type filter in the Reporting REST APIREST APIA web API design style where clients interact with resources over HTTP using standard methods like GET, POST, and DELETE, giving cyber tools a predictable, language-agnostic way to integrate with external services. Audit Log report enables retrieval of audit records by specific user action type.
New Audit Log report available in the Reporting REST API for querying user activity from Veracode Analytics.
Multiple domains can now be mapped to a single identity provideridentity providerA service that authenticates users and supplies identity information to other applications. It is the sign-in authority behind many single sign-on systems. for SP-initiatedSP-initiatedA SAML or OAuth/OIDC login flow where the Service Provider redirects the user to an Identity Provider to authenticate, rather than the IdP pushing an assertion unprompted; tools use it to trigger SSO from their own login page.SAML SSOSAML SSOAn authentication protocol that lets a user log in once with an identity provider and access multiple services without re-entering credentials, by passing signed XML assertions between parties..
Customizable PDF report now displays SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself. Component License Risk policy compliance in the Executive Summary and Detailed SCA Findings sections.
Customer Managed Encryption Keys (CMEK) allows encryption of Veracode Platform data using your own root encryption key managed in your KMSKMSA managed service that creates, stores, rotates and controls access to encryption keys, usually backed by HSMs. Applications ask it to encrypt or decrypt rather than holding keys themselves..
Administrators can now customize columns in the All Users table on the Administration page.
Upgraded Veracode Analytics to Looker 26.6, adding finer widget size granularity, persistent filtering, and tabbed dashboards.
23 details in Everything
└──▷ BREAKING ON UPGRADE
!The /healthcheck/status endpoint has been removed and now returns a 404 error; use the v2/principal endpoint in the Identity REST API to check authenticated connectivity instead.
!Success events have been removed from the Firewall dataset in Firewall Analytics.
Exempt health-check and metrics endpoints from country-based blocking so monitoring systems in blocked regions still reach them.
ini
COUNTRY_IGNORE_URI=/health /metrics\?.*
Send users to a welcome page after passing an antibot challenge instead of replaying their original (potentially POST) request.
ini
ANTIBOT_SUCCESS_URI=/welcome
Adds USE_LIMIT_REQ_GLOBAL and LIMIT_REQ_GLOBAL_RATE settings to cap the total aggregate request rate for a service across all clients and URLs, evaluated before per-IP and per-URL limits.
Adds COUNTRY_IGNORE_URI to skip country whitelist and blacklist checks for URIs matching a list of PCRE expressions, matched against both the request path and the full URI including query string.
Adds ANTIBOT_SUCCESS_URI to redirect clients to a fixed URI after successfully completing a challenge, rather than returning them to the originally requested page.
Audit, review, and clean up your Claude Code standing permissions.
GrantGuard v0.1.0 launches: audit and clean up risky Claude Code standing permissions from the CLI.
└──▷ GET THIS VERSION
$ git clone --branch v0.1.0 https://github.com/OpenVanta/GrantGuard.git
# already have the repo? check out this version:$ git checkout v0.1.0
└──▷ TRY IT
Quickly audit your personal Claude Code permissions for high-risk grants — inline API keys, credential-store reads, destructive wildcards — without modifying anything.
$ uv run grantguard.py audit
Open the interactive browser UI to review and selectively remove flagged permissions across a specific repo's Claude settings.
$ uv run grantguard.py ui /path/to/repo
New audit command scans Claude Code settings files and classifies every allow rule by risk — inline credentials, credential-store reads, destructive wildcards, unprompted remote pushes, and overbroad wildcards.
New ui command opens a local browser UI for interactive review and one-click removal of flagged permission grants.
Supports --scan for shallow discovery of .claude/settings*.json under target directories, and --deep-scan for broad recursive discovery.
New --fix flag on audit applies removals non-interactively, rewriting only the flagged rules while leaving all other settings intact.
New --tolerance flag (default or permissive) controls whether overbroad wildcard rules are flagged or treated as safe.
+3 moreshow less
New --show-safe flag on audit prints kept (non-flagged) rules alongside flagged ones for full visibility.
Reads all standard Claude Code permission sources by default: ~/.claude/settings.json, ~/.claude/settings.local.json, ~/.claude.json, and the platform managed settings file.
Cross-platform support for macOS, Linux, and Windows with zero third-party runtime dependencies.
Sublime Security is an email security platform that detects and blocks phishing, malware, and other email threats using customizable, logic-based detection rules.
Sublime Security adds Advanced Graymail Protection (Beta) with ML-based detection, user preference learning, and automated remediation for M365 and Google Workspace.
└──▷ HOW TO FIND IT
Enable graymail user preference learning so Sublime personalizes bulk-mail handling per user based on their folder moves.
📍In the console, go to Admin → Account and enable User Preferences.
Activate automated graymail remediation to automatically route and manage detected graymail without manual triage.
📍In the console, activate the Rule Fuzzy Attack Score: Advanced Graymail Detection and the Automation Remediate graymail flagged messages.
New dedicated Graymail view separates high-volume non-malicious mail (marketing, newsletters, subscriptions) from other detections across Microsoft 365 and Google Workspace.
Adds bi-directional User Preferences: Sublime learns per-user sender preferences when users move messages into or out of Promotions or Junk/Spam and applies them to future messages.
Introduces Fuzzy Attack Score, a new ML model that identifies unwanted bulk mail without relying on rigid signatures or sender-specific rules.
New Graymail Automation enables automatic routing, triaging, and management of detected graymail via the included Remediate graymail flagged messages automation.
New built-in detection rule Fuzzy Attack Score: Advanced Graymail Detection is included and auto-activated for orgs with Automatically Activate New Rules enabled.
+1 moreshow less
Revamped navigation replaces the Flagged view with Needs Remediation (closest equivalent) and adds Any Rule Match to review all messages matching rules and filters.
6 details in Everything
└──▷ BREAKING ON UPGRADE
!The Flagged view has been replaced; Needs Remediation is its closest equivalent for reviewing messages awaiting action.
Sublime Security adds four REST APIREST APIA web API design style where clients interact with resources over HTTP using standard methods like GET, POST, and DELETE, giving cyber tools a predictable, language-agnostic way to integrate with external services. endpoints to create, list, and manage child organizations and their licenses at scale.
New GET /v0/organizations/mine/child-organizations endpoint lists all child organizations with license and usage data, supporting status filtering, name search, pagination, and depth traversal.
New POST /v0/organizations/mine/child-organizations endpoint creates child organizations and optionally assigns licenses in a single API call.
New PUT /v0/organizations/mine/child-organizations/{id}/licensing endpoint updates the license on an individual child organization.
New bulk licensing endpoint applies license transitions to up to 200 child organizations at once, enabling automated renewals, upgrades, and offboarding.
Vectra AI is a cybersecurity platform that uses AI-driven threat detection and response to identify and prioritize attacker behaviors across hybrid cloud environments.
Vectra AI v9.13 adds expanded detections for Windows post-exploitation, HTTP attack tools, covert tunneling, credential theft, and cloud exfiltration.
New detections for SMBSMBA network file-sharing protocol developed by Microsoft that lets systems share files, printers, and named pipes over a network; cyber tools target or speak it because it is ubiquitous in Windows environments.-based lateral movementlateral movementMoving from an initially compromised system or identity to other resources in a network. Attackers often use stolen credentials or trusted connections to expand their access., malicious named pipe activity, and Cobalt Strike SMB beaconing across Windows environments.
Expanded HTTP-based detection coverage for malicious downloads, uploads, certificate abuse, and offensive framework activity using trusted Windows utilities.
New detection coverage for tunneling frameworks and covert channels, including web-based tunnels, public tunneling services, and custom BackConnect protocols.
New detections for Active DirectoryActive DirectoryMicrosoft's directory service for Windows networks that centralises authentication and authorisation; cyber tools target or integrate with it because it controls identity and access across most enterprise environments. Certificate Services abuse, suspicious AD database file transfers, and cloud storage exfiltration via common sync tools.
Vectra AI v9.13 expands ERSPANERSPANA Cisco-originated GRE-based tunneling protocol that encapsulates mirrored network traffic and ships it across Layer 3 to a remote capture point, letting cyber tools analyze traffic from distant segments without local tap hardware. support with 8- and 16-core virtual Sensor options and improved encapsulated traffic performance.
Supports 8- and 16-core virtual Sensors for ERSPAN-based monitoring deployments.
Improved configuration handling for ERSPAN capture interface addressing simplifies virtual and physical Sensor setup.
Performance optimizations increase processing efficiency for encapsulated traffic workloads in ERSPAN environments.
OpenShell is the safe, private runtime for autonomous AI agents.
OpenShell v0.0.86 adds AWS STSAWS STSAWS Security Token Service, an Amazon Web service that issues short-lived credentials for IAM roles, letting tools authenticate to AWS APIs without storing long-term secrets. AssumeRole refresh, network egress middleware, and TUITUIA text-based user interface that runs inside a terminal, using characters and ANSI escape codes to render menus, panes, and widgets without a graphical desktop, letting cyber tools ship rich interactivity with no GUI dependency. panel navigation.
└──▷ GET THIS VERSION
$ git clone --branch v0.0.86 https://github.com/NVIDIA/OpenShell.git
# already have the repo? check out this version:$ git checkout v0.0.86
Adds 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. AssumeRole refresh strategy and aws-s3 profile support for cloud provider integrations.
Prowler is the world’s most widely used open-source cloud security platform that automates security and compliance across any cloud environment.
Prowler 5.35.0 adds Lighthouse AI side chat and actions, one-step AWS Org onboarding, scan exclusions, and a redesigned sidebar.
└──▷ GET THIS VERSION
$ git clone --branch 5.35.0 https://github.com/prowler-cloud/prowler.git
# already have the repo? check out this version:$ git checkout 5.35.0
Adds Lighthouse AI side panel: open a context-aware AI chat from anywhere in the app without leaving the current page, with tabs to switch between finding details and the chat (Prowler Cloud/Private Cloud).
Enables Lighthouse AI to take actions: connect/remove providers, trigger and schedule scans, update scan settings, and manage mute rules via chat, gated by 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. (Prowler Cloud/Private Cloud).
Adds one-step AWS Organizations onboarding: a single CloudFormationCloudFormationAn AWS service that lets you describe and provision cloud infrastructure as declarative YAML or JSON templates, so a cyber tool can reproducibly spin up or tear down its entire AWS environment with a single API call. quick-create link deploys the management account role and a service-managed StackSetStackSetAn AWS CloudFormation construct that deploys and manages stacks across multiple accounts and regions from a single template, letting cyber tools provision infrastructure at scale without per-account manual work. to every member account (Prowler Cloud/Private Cloud).
Adds excluded_checks and excluded_services fields to scan configurations, allowing entire checks or services to be skipped per provider for faster, lower-noise scans.
Introduces a redesigned sidebar with grouped sections for security, settings, and help, collapsible configuration entries, route-aware active states, and a responsive mobile overlay.
+2 moreshow less
Renames core Prowler 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. tools from the prowler_app_* prefix to the shorter prowler_* namespace; legacy prowler_app_* names remain functional.
Enables dynamic providers to be renamed and deleted directly from the Providers table.
A vulnerability scanner for container images and filesystems
Grype v0.116.0 adds Go symbol-based reachability analysis, Ubuntu ESM support, architecture-aware matching, and Chainguard OSVOSVOpen Source Vulnerabilities, a vulnerability database and schema maintained by Google that gives cyber tools a machine-readable, cross-ecosystem format for querying and sharing CVE-style advisories. integration.
└──▷ GET THIS VERSION
$ git clone --branch v0.116.0 https://github.com/anchore/grype.git
# already have the repo? check out this version:$ git checkout v0.116.0
Adds lightweight Go symbol reachability analysis to suppress false positives for stdlib, golang.org/x, and third-party Go vulnerability advisories — only flags binaries that actually import the vulnerable symbols.
Deduplicates Go matches that are aliases of each other (same CVE reported under both a govulndb GO-* ID and its GHSA).
Adds a Chainguard OSV transformer for processing Chainguard vulnerability data.
Duplicates RHSA advisories across all applicable RHELRHELRed Hat Enterprise Linux, a commercially supported Linux distribution maintained by Red Hat, chosen by cyber tools for its long support lifecycle, certified hardware compatibility, and enterprise deployment prevalence. minor versions for more complete RHEL coverage.
Huntress is a managed cybersecurity platform that detects, investigates, and remediates threats targeting small and mid-sized businesses through persistent footholds and endpoint monitoring.
Managed ITDRITDRIdentity Threat Detection and Response: a discipline focused on detecting and responding to attacks that abuse identity systems such as Active Directory or IAM platforms, giving cyber tools visibility into credential-based intrusions. now supports IP- and CIDRCIDRA notation for a block of IP addresses, such as 10.0.0.0/24. The number after the slash says how many leading address bits identify the network.-based Expected Unwanted Access rules to suppress known-good login escalations.
└──▷ HOW TO FIND IT
Suppress login escalations from a known office or remote-worker IP range without creating overly broad country or VPNVPNA virtual private network, an encrypted tunnel between devices or networks. It carries traffic across another network as though the endpoints had a private connection. exceptions.
📍In the Huntress console, go to Managed ITDR › Expected Unwanted Access rules and create a new rule scoped to a single IP address or CIDR range.
Tenable Attack Surface Management discovers and monitors an organization's internet-facing assets, identifying exposures and vulnerabilities across the external attack surface.
Tenable ASMASMAttack Surface Management: the continuous practice of discovering, inventorying, and monitoring an organization's externally exposed assets so that cyber tools can flag unknown or misconfigured entry points before attackers find them. adds a built-in subscription to identify external-facing Microsoft SharePoint Server assets.
└──▷ HOW TO FIND IT
Identify all external-facing SharePoint Server instances in your attack surface to assess exposure to SharePoint-targeting CVEs.
📍In the Tenable ASM console, navigate to Subscriptions and enable the built-in subscription labeled Microsoft Sharepoint Server - v1.
Immuta is a data security platform that automates access control, privacy protection, and policy enforcement across cloud data environments.
Immuta adds managed groups for system-wide policies, automated Redshift discovery, OAuth 2.0OAuth 2.0An open authorization framework (IETF RFC 6749) that lets a service grant limited API access to a third party without sharing credentials, giving cyber tools a standard way to authenticate against external platforms. for Alation, and Delta Sharing row/column enforcement.
└──▷ HOW TO FIND IT
Write a subscription policy that allows every registered identity (agents and users) to subscribe to non-sensitive data sources, without maintaining manual group membership.
📍In the Immuta UI, go to Policies › Subscription Policies › New Policy, set the subject to the built-in group All Identities, and configure the access condition for your non-sensitive data sources.
Automated sensitive data discovery now available for Immuta's Amazon Redshift viewless integration, supporting regex, dictionary, and column-name-regex identifiers.
Row filtering and column masking now enforced for Databricks Delta Sharing recipients on Unity Catalog managed, external, and foreign tables.
Manual data policy resync in Snowflake data source health checks now resyncs both data policies and subscription policies.
Integration error banners replaced by error visibility through the new connections experience.
Snowflake back-end architecture update improves efficiency and performance at higher scale for the Snowflake integration.
+2 moreshow less
Documentation embedded directly in the Immuta UI with AI-powered chatbot accessible from the top menu.
Improved query performance for Snowflake policies that include advanced masking exceptions via a more efficient comparison method.
7 details in Everything
└──▷ BREAKING ON UPGRADE
!Immuta policy role names used to grant users access to Snowflake data sources will be renamed as part of the Snowflake architecture update beginning July 23, 2026; existing policy roles will be recreated under new names.
!When a data source is disabled in Immuta, Immuta-managed grants for data owners in the data platform will be revoked (behavior change release: on by default 8/1, enabled for all 9/1).
Ping Identity is an identity security platform that provides authentication, single sign-on, multi-factor authentication, and access management solutions to protect users and resources.
Ping Identity AIC adds optional custom report parameters, ordered promotions API results, and multi-line journey annotations.
Supports marking parameters in custom reports as optional, enabling more flexible report execution.
Promotions API now returns report metadata ordered by creation date, most recent first.
Journey annotation notes now render multi-line formatting in view mode, consistent with edit mode.
Keycloak is an open-source identity and access management solution that provides single sign-on, user authentication, and authorization for applications and services.
Keycloak 26.7 adds stateless mode preview, moving auth sessions and brute-force counters to the DB to enable simpler multi-cluster deployments.
└──▷ TRY IT
Start a single Keycloak node (or cluster) with stateless mode enabled to persist auth sessions and brute-force counters in PostgreSQL.
New stateless preview feature stores authentication sessions, action tokens, and brute-force counters in the database instead of distributed Infinispan caches.
Enables Multi-Cluster v2: two or more independent Keycloak clusters connected only through a synchronously replicated database, with no external Infinispan cluster required.
Cross-cluster cache invalidation now uses a database outbox/queuing table with a default polling interval of 100 ms, eliminating direct network connections between clusters.
Authentication sessions survive full cluster restarts during minor upgrades, keeping users mid-login from losing progress.
Multi-cluster failover is now environment-agnostic — not tied to Kubernetes or a specific cloud provider — using any synchronously replicated database (tested with Amazon Aurora PostgreSQL) and an external load balancer (tested with AWS Global Accelerator).
+1 moreshow less
Removes dependency on external Infinispan cluster for multi-site setups, eliminating its monitoring and failover/failback operational overhead.
BeyondTrust is a privileged access management platform that secures, controls, and monitors privileged accounts, credentials, and remote access across enterprise environments.
BeyondTrust Base 8.3.0 brings 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.-structured syslog, a unified patch inventory view, and upload progress visibility.
Remote syslog output now follows OCSF structure, enabling richer, standardized event forwarding to SIEMSIEMSecurity information and event management, collecting logs from many systems and correlating events to find suspicious activity. Analysts use it to investigate alerts and incidents. platforms with new log categories for on-premises appliances.
Local file uploads now show an estimated progress bar instead of a generic spinner, giving clearer visibility into upload duration.
Removes the opt-in default certificate mechanism, simplifying certificate configuration.
3 details in Everything
└──▷ BREAKING ON UPGRADE
!The ability to opt out of seamless (no-downtime) critical security updates is removed; these updates now apply automatically on on-premises appliances.
!If an appliance has not successfully communicated with the update server within a defined time window, key functionality is degraded or disabled.
BeyondTrust Remote Support 26.2.1 adds multi-group Jump Client installs, FIDO2FIDO2The open standard behind passkeys and hardware security keys, where a website checks a signature from the user's device instead of a password. Because the key is bound to the site, phishing pages cannot use it. Group Policy enforcement, AI session summaries, syslog streaming, and ARMARMA family of RISC processor architectures designed by Arm Holdings and licensed to chip makers worldwide; tools target it to run natively on mobile, embedded, and increasingly server and desktop hardware.-native support.
└──▷ TRY IT
Deploy a Jump Client into multiple Jump Groups simultaneously at install time, avoiding post-deployment API calls to reassign group membership.
Assigns a Jump Client to multiple Jump Groups at installation time via a comma-separated JC_JUMP_GROUP installer parameter, eliminating post-deployment API calls to reassign group membership.
Enforces FIDO2 phishing-resistant 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. at the Group Policy level, using the same model as existing TOTPTOTPA one-time password algorithm (RFC 6238) that generates short-lived codes from a shared secret and the current time, giving cyber tools phishing-resistant second-factor authentication without a server round-trip. enforcement; FIDO2 is also now supported on Pathfinder deployments.
Adds search, filtering, CSV export, and API access to the Duplicate Jump Clients report for programmatic duplicate detection and resolution at scale.
Automatically generates AI-powered session summaries on session end, including case summary, step-by-step resolution, system tags, and key highlights; summaries are editable, exportable in multiple formats, multi-language, and integrate with Salesforce and ServiceNow.
Brings vault discovery and managed/shadow account capabilities natively into the product, including a filter dropdown on the Vault > Endpoints page, new import and auto-create options for SSH, RDPRDPA Microsoft protocol that lets a user view and control a remote Windows desktop over a network, giving cyber tools a standard channel to audit, attack-simulate, or monitor Windows systems remotely., and Password Safe Jump Items, and a direct-access Password Safe Discovery method.
+8 moreshow less
Adds a Collaboration Mode shortcut in the Representative Console to start a screen-sharing session with yourself and immediately invite other representatives, replacing the legacy Presentation feature.
Supports vault credential injection during an active RDP session (run-as scenario) without disconnecting, requiring the BeyondTrust RDP Tools component on the endpoint.
Streams appliance platform and infrastructure logs (web server access/error, authentication, SSLSSLA cryptographic protocol for encrypting network connections, now deprecated in favor of TLS but whose name persists colloquially. Cyber tools reference it when handling HTTPS or other encrypted-channel inspection and termination./HAProxy, proto-handler, and more) to an existing syslog destination for SIEMSIEMSecurity information and event management, collecting logs from many systems and correlating events to find suspicious activity. Analysts use it to investigate alerts and incidents. integration; available for on-premises deployments.
Officially supports the Jump Client on Windows Server Core (2019, 2022, and Azure-hosted), including remote shell, file transfer, session recording, credential injection, and auto-update.
Runs BeyondTrust clients natively on ARM-based systems (e.g., Windows 11 ARM) without x86 emulation, covering credential injection, Virtual Smart Card, and Jump Client download via API.
Supports vault credential auto-entry for Remote VNC Jump sessions (Cloud only), enabled per asset via the Config API with a recorded consent notice, useful for OT/ICSICSIndustrial Control Systems — hardware and software that monitor and control physical processes like power grids, pipelines, and manufacturing lines, making them a target for cyber tools focused on operational technology environments. legacy endpoints.
Allows representatives to pop out each remote monitor into its own independent, resizable window instead of cycling through monitors one at a time.
Scales the Vault to support up to 200,000 accounts.
13 details in Everything
└──▷ BREAKING ON UPGRADE
!The ability to opt out of seamless (no-downtime) critical security updates is removed; these updates are now applied automatically on upgrade.
!RADIUSRADIUSA networking protocol for centralised authentication, authorisation, and accounting of users connecting to a network, maintained under IETF standards. Cyber tools use it to integrate with existing enterprise identity infrastructure. authentication provider is deprecated: existing configurations can be edited but new ones cannot be created — migrate to SAMLSAMLA standard for passing signed identity assertions between a sign-in provider and an application. It lets a user sign in through their organization's identity provider. or LDAPLDAPA protocol for querying and modifying directory services, standardized under RFC 4511, commonly used to look up users, groups, and attributes in corporate identity stores like Active Directory. before future versions remove the provider entirely.
SailPoint is an identity security platform that manages and governs user access, entitlements, and permissions across enterprise applications and systems.
SailPoint previews next-gen mainframe connectors for RACF, ACF2, and Top Secret that eliminate the Connector Gateway layer.
Introduces direct connectivity from SailPoint Identity Management products to the SailPoint agent on the mainframe, removing the requirement to deploy and maintain a Connector Gateway.
Adds new security and maintainability improvements to the RACF, ACF2, and Top Secret integrations on z/OS.
Includes migration tooling to help transition existing configurations and customizations from the legacy connector to the new architecture.
Okta Identity Engine is a platform that enables developers to build flexible, customizable authentication and identity management workflows into their applications.
Okta adds cross-app access for AI agents, GA agent-to-agent connections, and new device OS account endpoints.
Enables cross-app access for AI agents and applications.
Agent-to-agent connections reaches General Availability in Production.
New endpoints for device OS accounts now available in Early Access (Preview).
RegScale is a continuous compliance automation platform that helps organizations manage security controls, audits, and regulatory requirements in real time.
RegScale 6.37.39 adds eMASSeMASSA U.S. DoD web application that manages the Risk Management Framework process, tracking system authorizations, controls, and POA&Ms so cyber tools can automate compliance reporting into a single authoritative record. asset inventory sync and expanded stakeholder address fields in the API.
New eMASS API hardware and software baseline sync commands pull asset inventory from eMASS, create or update RegScale assets, and push new assets back to eMASS.
Kiro is an AI-powered IDE by AWS that uses "specs" to turn requirements into code through structured, agentic software development workflows.
Kiro CLI v3 gains an Introspect subagent for in-line feature guidance and global hooks that fire across all workspaces.
└──▷ TRY IT
Ask the Introspect subagent how to configure a custom hook without leaving your terminal.
$ kiro-cli --v3
New Introspect subagent answers questions about Kiro's features, explains custom agents, hooks, and steering files, and suggests workflow configurations — all inline (v3 only).
Global hooks defined in ~/.kiro/hooks/ now automatically apply to every workspace, eliminating per-project duplication for cross-cutting behaviors like linting on save or pre-commit security checks (v3 only).
Lightweight coding agent that runs in your terminal
Amazon Bedrock provider now supports custom transports: override base_url, auth, and HTTP headers alongside AWS profile and region.
└──▷ GET THIS VERSION
$ git clone --branch rust-v0.145.0-alpha.20 https://github.com/openai/codex.git
# already have the repo? check out this version:$ git checkout rust-v0.145.0-alpha.20
Enables custom base_url, auth, and http_headers overrides on the built-in amazon-bedrock provider for custom transport scenarios.
Supports command-based bearer authentication and configured endpoints for Bedrock without requiring AWS request signing.
Replaces the Bedrock account credentialSource enum with the usesCodexManagedCredentials boolean for consistent reporting of managed and externally authenticated configurations.
3 details in Everything
└──▷ BREAKING ON UPGRADE
!The Bedrock account credentialSource enum is replaced by the usesCodexManagedCredentials boolean; any configuration or code referencing credentialSource will break.
0 more releases in this issue2 more releases in this issue · 2026-07-16 → 2026-07-17
Codex CLI alpha.19 adds concurrent terminal session writes and an app-server metadata API
└──▷ GET THIS VERSION
$ git clone --branch rust-v0.145.0-alpha.19 https://github.com/openai/codex.git
# already have the repo? check out this version:$ git checkout rust-v0.145.0-alpha.19
Enables parallel write_stdin tool calls across independent terminal sessions, with per-session serialization to keep output buffers consistent.
Adds experimental app/read API for fetching metadata for up to 100 app IDs, with optional tool summaries, deduplication, caching, and partial-miss reporting.
Validates reasoning effort against model metadata after agent roles override model settings, catching unsupported combinations at spawn time.
Stabilizes the active-turn environment so in-progress turns are unaffected by settings updates during deferred execution.
Requires data: scheme URLs for code-mode image output, rejecting remote HTTP URLs and other malformed image sources.
Codex CLI alpha.18 unifies multi-agent config under agents, adds encrypted 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. content support, and improves skill-selection with BM25BM25The classic keyword-ranking formula behind traditional search engines. Hybrid search combines it with embedding similarity so exact terms and meaning both count., n-gram, and multi-query methods.
└──▷ GET THIS VERSION
$ git clone --branch rust-v0.145.0-alpha.18 https://github.com/openai/codex.git
# already have the repo? check out this version:$ git checkout rust-v0.145.0-alpha.18
Adds agents.enabled config key and unifies multi-agent settings (concurrent thread limit, subagent model, reasoning effort) under a single [agents] section, with resolved values visible in TUITUIA text-based user interface that runs inside a terminal, using characters and ANSI escape codes to render menus, panes, and widgets without a graphical desktop, letting cyber tools ship rich interactivity with no GUI dependency. debug-config output.
Adds agents.max_concurrent_threads_per_session (with agents.max_threads as an alias) to control the spawned-thread limit across both multi-agent backends.
Adds agent_default_subagent_model and agent_default_subagent_reasoning_effort config fields so spawned agents honor configured model defaults.
Exposes the agent_type override in the spawn_agent schema only when agent roles are configured, avoiding an inapplicable argument in the model-visible tool.
Preserves encrypted MCP tool output: text content marked with _meta["codex/encryptedContent"] = true is now converted to encrypted_content function-call output items.
+4 moreshow less
Adds cacheWriteInputTokens to the rawResponse/completed usage breakdown schema.
Adds fielded BM25 skill selection that ranks skills across name, short-description, and description fields with name-weighted and IDF scoring.
Adds character n-gram skill selection covering related word forms, typos, and CJK text without word boundaries.
Adds multi-query lexical skill selection that splits compound queries into sentence- and connector-delimited views and merges ranked candidates across views.
Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.
Claude Code v2.1.212 adds background /fork, auto-mode reset, per-session WebSearch/subagent caps, and auto-backgrounding slow 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. calls.
└──▷ GET THIS VERSION
$ git clone --branch v2.1.212 https://github.com/anthropics/claude-code.git
# already have the repo? check out this version:$ git checkout v2.1.212
└──▷ TRY IT
Reset auto-mode to its defaults non-interactively in a script or CI context.
$ claude auto-mode reset --yes
Cap WebSearch calls and subagent spawns to tighter limits for a long-running autonomous session to prevent runaway loops.
$ CLAUDE_CODE_MAX_WEB_SEARCHES_PER_SESSION=50 CLAUDE_CODE_MAX_SUBAGENTS_PER_SESSION=20 claude
Lower the MCP auto-background threshold to 30 seconds so slow MCP calls yield the session sooner.
$ CLAUDE_CODE_MCP_AUTO_BACKGROUND_MS=30000 claude
Redesigns /fork to copy the current conversation into a new background session (visible in claude agents) rather than an in-session subagent; the old in-session subagent behavior is now /subtask.
Adds claude auto-mode reset command to restore default auto-mode configuration, with an optional --yes flag to skip the confirmation prompt.
Adds a per-session WebSearch call cap (default 200, tunable via CLAUDE_CODE_MAX_WEB_SEARCHES_PER_SESSION) to prevent runaway search loops.
Adds a per-session subagent spawn cap (default 200, override via CLAUDE_CODE_MAX_SUBAGENTS_PER_SESSION); /clear resets the budget.
MCP tool calls exceeding 2 minutes automatically move to the background to keep the session responsive; threshold is configurable or disableable via CLAUDE_CODE_MCP_AUTO_BACKGROUND_MS.
+8 moreshow less
Typing /resume in the agent view now opens a session picker — including previously deleted sessions — and resumes the chosen session as a background session.
Improves prompt caching: mid-conversation system block now works behind 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. gateways and custom base URLs (Bedrock, Vertex, 1P).
Reduces token usage in inter-agent messaging by no longer duplicating SendMessage bodies into replayed history and tool results.
Extends Enterprise forceLoginMethod enforcement to VS Code extension, SDK, setup-token, and install-github-app logins.
Headless/SDK sessions now apply a set_model control request mid-turn so the next LLM round-trip uses the new model immediately.
Agent view and claude agents --json now show sessions waiting on a sandbox, MCP-input, or managed-settings prompt as Needs input instead of Working.
Session transcripts now record the reasoning effort level on each assistant message.
Background agent cold-attach now instantly renders the formatted transcript while the session boots, eliminating the blank-wait screen.
13 details in Everything
└──▷ BREAKING ON UPGRADE
!The Task tool's mode parameter is now ignored; subagents inherit the parent session's permission mode by default.
Pi v0.80.9 adds Kimi K3 support with deferred tool loading across five providers and switches xAI default to Grok 4.5.
└──▷ GET THIS VERSION
$ git clone --branch v0.80.9 https://github.com/earendil-works/pi.git
# already have the repo? check out this version:$ git checkout v0.80.9
Adds Kimi K3 model support across Kimi Coding, Moonshot AI, Moonshot AI China, OpenRouterOpenRouterA unified API gateway that routes requests to dozens of LLM providers under one endpoint, letting cyber tools swap or compare models without rewriting provider-specific integration code., and Vercel AI Gateway.
Adds Kimi deferred tool loading for progressive, extension-driven tool activation via Kimi's native protocol.
Changes the default xAI model to Grok 4.5 and updates xAI login to use a prefilled device-authorization link labeled Sign in with SuperGrok or X Premium.
3 details in Everything
└──▷ BREAKING ON UPGRADE
!Grok 3, Grok 3 Fast, Grok 4.20 variants, and Grok Code Fast 1 have been removed from the built-in xAI model catalog — any configuration referencing these models will break.
Framework for orchestrating role-playing, autonomous AI agents. By fostering collaborative intelligence, CrewAI empowers agents to work together seamlessly, tackling complex tasks.
CrewAI 1.15.4 promotes the Skills Repository to stable, making AI coding-agent guidance an officially supported feature.
└──▷ GET THIS VERSION
$ git clone --branch 1.15.4 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:$ git checkout 1.15.4
└──▷ TRY IT
Install the now-stable CrewAI Skills into Claude Code so it automatically applies CrewAI best practices when you scaffold Flows, configure Crews, or design agents.
CrewAI 1.15.3 adds step interception hooks with @on, org ID support in PlusAPI, and declarative Flow execution in the TUITUIA text-based user interface that runs inside a terminal, using characters and ANSI escape codes to render menus, panes, and widgets without a graphical desktop, letting cyber tools ship rich interactivity with no GUI dependency..
└──▷ GET THIS VERSION
$ git clone --branch 1.15.3 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:$ git checkout 1.15.3
└──▷ USE IT
Intercept execution boundaries in a crew or flow to log, audit, or gate agent activity at specific lifecycle points.
python
from crewai import on
@on("before_llm_call")
def audit_llm_input(context):
print("LLM about to be called with:", context)
@on("after_llm_call")
def log_llm_output(context):
print("LLM responded with:", context)
Adds step interception points and a generic hook dispatcher, enabling fine-grained execution-boundary callbacks via @on decorators.
Adds organization_id parameter to the PlusAPI client for multi-org deployments.
Supports running declarative Flows on the TUI with headless terminal fallback.
CrewAI 1.15.3a1 adds step interception hooks via @on, declarative flow support in the TUITUIA text-based user interface that runs inside a terminal, using characters and ANSI escape codes to render menus, panes, and widgets without a graphical desktop, letting cyber tools ship rich interactivity with no GUI dependency., and an organization ID parameter for the PlusAPI client.
└──▷ GET THIS VERSION
$ git clone --branch 1.15.3a1 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:$ git checkout 1.15.3a1
Adds step interception points and a generic hook dispatcher, enabling @on-decorated callbacks at execution boundaries for fine-grained observability and control.
Supports running declarative flows on the TUI with a headless terminal fallback.
Adds an organization ID parameter to the PlusAPI client for multi-org deployments.
Improves custom OpenAI URL handling for self-hosted or proxy 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. endpoints.
Get up and running with Kimi-K2.6, GLM-5.1, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.
Ollama v0.32.1 adds CWD-aware agent context, actionable web-auth guidance, and MLXMLXApple's machine-learning framework for Apple silicon, which uses the unified memory shared by CPU and GPU to run and fine-tune models on a Mac. load-timeout support.
└──▷ GET THIS VERSION
$ git clone --branch v0.32.1 https://github.com/ollama/ollama.git
# already have the repo? check out this version:$ git checkout v0.32.1
Adds current working directory to the interactive agent's system prompt for better project context.
Agent web search and fetch now surface an actionable ollama signin prompt when authentication is required, instead of a generic error.
MLX text model loading now respects the OLLAMA_LOAD_TIMEOUT environment variable.
Improves Gemma 4 tool callingtool callingA model's ability to answer with a structured request to run a named function or API, which the application executes and returns. It is how agents take actions. and multi-turn reasoning, including more reliable tool-response continuations.
THE CYBER TOOLCHAIN