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.
StackHawk is a dynamic application security testing (DAST) tool that helps developers find and fix security vulnerabilities in APIs and web applications during development.
StackHawk adds self-upgrade, scan drill-down commands, standalone binary, AI skills status, and org-wide Findings page.
└──▷ TRY IT
Check whether a newer HawkScan version is available before committing to an upgrade in a CI pre-flight step.
$ hawk upgrade --check
Iterate on an auth script against a live staging URL without running a full scan, with auto-reload on save.
Inspect the exact attack surfaceattack surfaceThe exposed systems, interfaces and access points an attacker could try to use. It includes internet-facing services as well as identities, software and other reachable resources. (including GraphQL operations) exercised by a completed scan to verify coverage.
$ hawk op scan uris --app my-api --graphql
New hawk upgrade command updates HawkScan CLI in place, detecting install method (Homebrew, binary, MSI, PKG, Docker) and printing the exact command when direct upgrade isn't possible; --check previews available versions and --version <x.y.z> pins to a specific release.
New hawk skills status command reports which StackHawk AI agent skills (Claude Code, Codex, Copilot) are installed and whether they are current, fetching latest published versions live.
New hawk op scan config retrieves the exact effective configuration a completed scan ran with; hawk op scan uris lists the full attack surface including GraphQL operations and JSON-RPCRPCRemote Procedure Call, a protocol that lets a program execute a function on another machine as if it were local, giving cyber tools a structured way to communicate across processes or hosts without managing raw sockets. methods.
--app flag on all hawk op commands now accepts either an application name or a UUIDUUIDA universally unique identifier, usually a 128-bit value used to identify an object across systems. It is designed to make independently generated identifiers unlikely to collide., eliminating the need to look up UUIDsUUIDsA universally unique identifier, usually a 128-bit value used to identify an object across systems. It is designed to make independently generated identifiers unlikely to collide. separately.
New hawk validate script command runs HawkScript auth logic against a single target URL without a full scan; --watch enables daemon mode that re-executes on every file save.
+12 moreshow less
HawkScan 6.0.0 ships as a self-contained native binary for macOS (Apple Silicon/Intel), Linux (x86_64/arm64), and Windows (x64/arm64) — no JDK or JRE required.
hawk op is now fully integrated into the hawk binary, eliminating the separately installed hawkop binary for platform operations.
HawkScan 6.0.0 automatically trusts the OS native certificate store (macOS Keychain, Windows Certificate Store, Linux system PEM bundle), enabling transparent TLS-intercepting proxies (Zscaler, Netskope, corporate CAs) with zero configuration; disable with --no-native-truststore.
New --browser flag for hawk init skips all TTY prompts, opens the browser immediately, and always prints the authentication URL to stdout — designed for AI coding agents.
New hawk flightpath command runs the auth-authoring loop entirely locally, driving a browser-based login to produce a validated stackhawk-auth.yml without cloud infrastructure.
New hawk config subcommands (hawk config list, hawk config show <path>, hawk config recipe <path>) surface HawkScan's full configuration reference from the CLI; --text outputs raw markdown.
Primary-host matching is now exact (normalized scheme, default ports stripped), and hosted-scanner HARHARAn HTTP Archive file, a JSON-format log of all network requests and responses captured by a browser's developer tools, used by cyber tools to replay, inspect, or analyze web traffic without re-running a live session. uploads are scoped to the primary host only, excluding CDNCDNA content delivery network, servers distributed across locations that cache and serve content near readers. It can reduce loading time and traffic to the original server. and third-party traffic by default.
New StackHawk Storage Domains (scan-results.prod.stackhawk.com and hosted-assets.prod.stackhawk.com) front platform object storage via CloudFront with short-lived presigned URLs.
New Findings page provides a single org-wide table to view, sort, and filter all open findings across every app and environment, refreshed automatically after each scan.
New Service Keys provide long-lived, organization-owned API keys for CI/CD scanning with a Scan Only role, managed under Settings → Service Keys.
FlightPath (private preview) automatically generates authenticated scan configurations by capturing an interactive browser login session using an AI agent — no YAML required.
StackHawk CircleCI Orb v2.0.0 upgrades to orb-tools v12 build tooling.
17 details in Everything
└──▷ BREAKING ON UPGRADE
!The hawk-{version}.zip download format is removed; pipelines downloading the ZIP must migrate to the OS/architecture-specific binary at https://download.stackhawk.com/hawk/{VERSION}/{PLATFORM}/hawk (Windows: hawk.exe).
!StackHawk CircleCI Orb v2.0.0: job and parameter names are renamed to snake_case — e.g., stackhawk/hawkscan-local becomes stackhawk/hawkscan_local and app-id becomes app_id. Pipelines pinned to a 1.x version are unaffected until upgraded.
StepSecurity is a platform that secures CI/CD pipelines by detecting and preventing supply chain attacks, hardening GitHub Actions workflows, and monitoring runtime behaviors.
StepSecurity adds a secure-registry CI/CD control and OIDCOIDCOpenID Connect, a sign-in protocol built on OAuth 2.0. An application uses identity information from a trusted provider instead of maintaining a separate password for each user. replaceability tracking for Actions secrets.
└──▷ HOW TO FIND IT
Identify CI/CD jobs pulling packages from public registries so you can route them through a secure proxy before a supply-chain compromise occurs.
📍In the StepSecurity console, go to Overview dashboard › Controls, open the Jobs Should Use a Secure Registry Instead of Public Package Registries control, then filter findings by repository or workflow to prioritize remediation.
Find secrets that can be eliminated entirely by switching to OIDC, reducing long-lived credential exposure across your repositories.
📍In the StepSecurity console, go to Actions Secret › Repository Secrets tab, then enable the OIDC Replaceable filter to list only secrets (e.g., NPM_TOKEN) that trusted publishing can replace.
New Overview dashboard control detects CI/CD jobs that download packages directly from public registries (registry.npmjs.org, pypi.org, registry-1.docker.io) instead of a secure proxy registry.
Control findings are filterable by repository, workflow, and compliance status, with remediation tracking and finding suppression on the control detail page.
Repository Secrets tab now surfaces Usage Status (Active, Stale, Unused, Unknown) for each secret, including which workflow file and step references it and the last run time and conclusion.
Secrets that can be replaced with OIDC (e.g., NPM_TOKEN replaceable via npm trusted publishing) are flagged in the Actions Secret section.
New summary cards on the Repository Secrets tab show Total Secrets, Unused Secrets, Stale Secrets, and OIDC Replaceable counts for an at-a-glance secrets posture view.
+1 moreshow less
Filtering on the Repository Secrets tab supports search by repository or secret name, status filter, and an OIDC-replaceable-only toggle.
Cloudflare is a cloud platform that provides website security, performance optimization, and DDoS protection by routing traffic through its global network.
Returns EDE 9 (DNSKEY Missing) alongside EDE 33 to expose the underlying DNSSECDNSSECA set of DNS extensions, maintained by the IETF, that add cryptographic signatures to DNS records so resolvers can verify responses have not been tampered with in transit. chain-of-trust failure that triggered the NTA.
Emits EDE 33 on every response under an active NTA — including non-DNSSEC domains — so the entire affected zone is transparently flagged.
Burp Suite is a web security testing platform that intercepts, analyzes, and manipulates HTTP/S traffic to identify vulnerabilities in web applications.
Burp Suite 2026.7 launches beta of AI-driven agentic pentesting that autonomously pursues user-defined testing goals.
ElcomSoft develops forensic and password recovery tools that help law enforcement, investigators, and security professionals access encrypted files, devices, and accounts.
Elcomsoft Phone Breaker 11.03 adds universal binary for macOS and restores SMSSMSShort Message Service, the protocol built into cellular networks for sending text messages between phones; cyber tools use it for alerting, two-factor authentication delivery, or phishing simulation over mobile channels.-based iCloud 2FA sign-in.
Ships as a universal binary, running natively on both Apple Silicon and Intel Macs without Rosetta translation.
Removes the pre-login 2FA status check that Apple no longer supports, enabling successful authentication flows.
Palo Alto Networks is a cybersecurity platform that provides advanced firewalls, cloud security, and AI-driven threat detection to protect networks and enterprises.
Prisma AIRS AI Gateway reaches GA with inline 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./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./A2AA2AAgent2Agent, an open protocol started by Google for AI agents from different vendors to discover each other, delegate tasks and exchange results. It complements MCP, which connects an agent to tools. enforcement, agent identity binding, and quantum-resilient SASESASEA network architecture model combining wide-area networking and network security functions into a single cloud-delivered service, letting cyber tools enforce consistent policy regardless of where users or workloads sit. fabric via AT&T partnership.
Introduces Prisma AIRS AI Gateway as a unified control plane sitting inline between AI apps, LLMsLLMsA 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., MCP tool calls, and A2A interactions for centralized governance.
Adds unified observability across all AI usage: tracking users, projects, models, token counts, latency, and cost in a single view to expose shadow AI.
Enables central governance of approved models, tools, and access with FinOps controls including per-team/project budget enforcement and rate limiting before access is granted.
Adds Coding Assistant Security that replaces raw provider API keys with scoped, per-user/team credentials to protect secrets and proprietary code.
Introduces Agent Identity Security that binds verifiable, ephemeral identities to agents at execution time, enforcing just-in-time least-privilege access.
+7 moreshow less
Delivers Runtime Security with inline prompt and response inspection aligned to OWASP LLM Top 10OWASP LLM Top 10A ranked list of the ten most critical security risks specific to large language model applications, maintained by OWASP, giving cyber tool developers a shared taxonomy for prioritizing LLM attack surfaces. and 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. Top 10 for Agentic Applications, blocking 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 data exfiltrationdata exfiltrationMoving data out of a system or organization without authorization. It can happen through network transfers, removable media or other channels..
Supports Universal API for traffic distribution across LLM providers with quota enforcement and failover.
Adds Post-Quantum CryptographyPost-Quantum CryptographyCryptography designed to stay secure against a large quantum computer, which could break today's RSA and elliptic-curve keys. NIST standardized the first algorithms in 2024 as ML-KEM, ML-DSA and SLH-DSA. (PQCPQCCryptography designed to stay secure against a large quantum computer, which could break today's RSA and elliptic-curve keys. NIST standardized the first algorithms in 2024 as ML-KEM, ML-DSA and SLH-DSA.) to Prisma SD-WANSD-WANSoftware-defined wide-area networking: an approach that decouples WAN routing logic from physical hardware, letting operators centrally control traffic paths across MPLS, broadband, and LTE links to reduce cost and improve resilience. via hybrid key exchanges based on IETF RFC 9370, 9242, and 8784, wrapping data in classical and quantum-resistantquantum-resistantCryptography designed to stay secure against a large quantum computer, which could break today's RSA and elliptic-curve keys. NIST standardized the first algorithms in 2024 as ML-KEM, ML-DSA and SLH-DSA. protection simultaneously.
Transitions all Prisma SASE control plane traffic to TLS 1.3 for quantum-resilient management plane protection.
Encrypts all network telemetry and metadata in transit via TLS 1.3 with region-geofenced storage and no cross-region aggregation.
Embeds crypto-agility natively so PQC algorithms can be updated via automated cloud updates without hardware replacement.
Adds Secure Boot support on Prisma SD-WAN ION hardware with TPMTPMA dedicated hardware chip, standardized by the Trusted Computing Group, that stores cryptographic keys and measurements in tamper-resistant silicon, giving cyber tools a root of trust that software alone cannot fake.-backed hardware root-of-trust, cryptographically signed boot chain, and PQC-secured management/control/data plane tunnels.
New provider: Scans E2E Networks with 27 checks covering compute nodes, networking, security groups, load balancers, block/file storage, and managed databases.
New AWS check ec2_ami_account_block_public_access: verifies AMI block public access is enabled at the account level per region.
New AWS check datapipeline_pipeline_no_secrets_in_definition: scans Data Pipeline object fields, parameter objects, and parameter values for hardcoded secrets using Kingfisher.
New AWS check elbv2_listener_pqc_tls_enabled: verifies ELBv2 HTTPSHTTPSHTTP carried over an encrypted TLS connection. It protects traffic in transit and lets a browser verify the server's certificate./TLS listeners use post-quantumpost-quantumCryptography designed to stay secure against a large quantum computer, which could break today's RSA and elliptic-curve keys. NIST standardized the first algorithms in 2024 as ML-KEM, ML-DSA and SLH-DSA. TLS security policies (TLS 1.2+) to reduce harvest-now-decrypt-later exposure.
New AWS check amplify_app_no_secrets_in_environment: scans Amplify app and branch environment variables and build settings (buildSpec) for hardcoded secrets using Kingfisher.
+6 moreshow less
New Azure check app_function_ensure_http_is_redirected_to_https: verifies that Function Apps enforce HTTPS-only traffic.
New Kubernetes check core_minimize_hostpath_volume_mounts: detects Pods that use hostPath volumes.
New Kubernetes check core_readonly_root_filesystem_enabled: verifies every container in each Pod explicitly sets readOnlyRootFilesystem: true in its security context.
New STACKIT check iaas_server_public_ip_attached: flags IaaS servers with a public IP address directly attached to a network interface.
New Cross-Provider Compliance view (Prowler Cloud/Private Cloud): rolls up the most recent scan of every compatible provider into a single compliance posture per framework (CISCISA nonprofit that publishes consensus-based configuration benchmarks and controls for hardening operating systems, cloud environments, and software; cyber tools use it as a ready-made compliance baseline. Controls 8.1, CSA CCM 4.0, DORA 2022/2554) with per-provider breakdown and combined executive PDF export.
UI now lists, filters, and renders dynamically registered (custom) providers with a generic icon, humanized label, and Custom badge when no bespoke assets exist.
11 details in Everything
└──▷ BREAKING ON UPGRADE
!UI integration enable flags renamed: UI_SENTRY_ENABLE → UI_SENTRY_ENABLED, UI_GOOGLE_TAG_MANAGER_ENABLE → UI_GOOGLE_TAG_MANAGER_ENABLED, UI_POSTHOG_ENABLE → UI_POSTHOG_ENABLED; deployments that set the former names must update them.
Sysdig Secure is a cloud-native security platform that detects threats, manages vulnerabilities, ensures compliance, and protects containers, Kubernetes, and cloud environments at runtime.
Sysdig Secure now supports multiple S3/S3-compatibleS3-compatibleAn object-storage API that mirrors Amazon S3's HTTP interface, so any client or cyber tool written for AWS S3 works unchanged against any compliant storage backend. custom storage buckets across Captures, Response Actions, and Rapid Response.
└──▷ HOW TO FIND IT
Connect a MinIO bucket to use as storage for Captures and Rapid Response sessions.
📍In the console, go to Integrations › Data Sources › Storage, click Add Bucket, select S3-compatible endpoint, enter your MinIO endpoint URL and credentials, then click Test Connection and Save.
Supports connecting multiple S3 or S3-compatible buckets (MinIO, Google Cloud Storage, and others) for use across Captures, Response Actions, and Rapid Response.
New Custom Storage configuration supports AWS access keys, AWS role delegation (AssumeRole), and S3-compatible endpoints, with connection testing on save and per-bucket failure history.
2 details in Everything
└──▷ BREAKING ON UPGRADE
!The previous single shared S3 Capture Storage configuration is replaced by Custom Storage; existing setups must be migrated to the new multi-bucket configuration under Integrations > Data Sources > Storage.
CLI tool and library for generating a Software Bill of Materials from container images and filesystems
Syft v1.48.0 adds vcpkg, macOS .app, KerberosKerberosA network authentication protocol that uses tickets issued by a trusted service. Users and services can prove their identities without sending a password to each application. 5, SafeTensor, and Ubuntu Pro/ESM cataloging plus Go binary symbol capture.
└──▷ GET THIS VERSION
$ git clone --branch v1.48.0 https://github.com/anchore/syft.git
# already have the repo? check out this version:$ git checkout v1.48.0
Adds dependency relationship edges from mix.lock to SBOMSBOMA software bill of materials, an inventory of the components and versions inside an application. It helps teams find affected software when a dependency has a vulnerability or licensing issue. output.
Adds SafeTensor model type cataloging to SBOM output for ML model inventories.
Captures symbols from Go binaries for richer binary-level SBOM data.
Detects Ubuntu Pro/ESM extended support status on scanned systems.
Adds a scan duration timer to Syft output.
+4 moreshow less
Adds support for vcpkg C/C++ package manager cataloging.
Adds a macOS .app bundle cataloger to inventory native macOS applications.
Exabeam is a security information and event management (SIEM) platform that uses AI and behavioral analytics to detect, investigate, and respond to cybersecurity threats.
Exabeam Threat Center adds caseClosedTimestamp field to case details, enabling Time-to-Close metrics.
Adds caseClosedTimestamp (date-time string) to the Get Case Details response schema, exposing when a case was closed for Time to Close (TTC) and Mean Time to Close (MTTC) calculations.
Google SecOps is a cloud-based security operations platform that helps organizations detect, investigate, and respond to cybersecurity threats using AI-powered analytics and threat intelligence.
Google SecOps dashboards gain Advanced Filtering with dynamic query variables, regex support, and multi-select logic injection into YARA-LYARA-LA rule language developed by Google for Chronicle that extends YARA syntax to express multi-event, time-windowed detections over log data rather than single-file pattern matching. at runtime.
└──▷ HOW TO FIND IT
When investigating across multiple threat actor IPs, use a multi-select advanced filter token to fan out a single YARA-L dashboard query across all selected values without editing the query manually.
📍In the dashboard editor, go to Filters › Add Advanced Filter, define a Token Variable (e.g., source_ip), set Filter Value Generation to a YARA-L query result or static list, enable Multi-Select, and set the delimiter to | — then reference the token in your YARA-L rule where the value should be injected at runtime.
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.
Supports token values generated dynamically from YARA-L query results or entered as a static manual list.
Enables customizable prefix/suffix wrappers around token values to express regex or other query logic.
Supports multi-select tokens with a configurable delimiter (e.g., |) for combining multiple values in a single YARA-L query.
Fleet 4.89.0 adds policy label targeting, setup experience policy checks, new vulnerability filtering, Google Workspace IdPIdPA service that authenticates users and supplies identity information to other applications. It is the sign-in authority behind many single sign-on systems. integration, and more.
└──▷ GET THIS VERSION
$ git clone --branch fleet-v4.89.0 https://github.com/fleetdm/fleet.git
# already have the repo? check out this version:$ git checkout fleet-v4.89.0
└──▷ TRY IT
Filter vulnerable software by CVSSCVSSCommon Vulnerability Scoring System, a method for describing a vulnerability's characteristics and severity. Its score is not the probability that attackers will exploit it. score and known exploit status on a device's self-service software tab — useful for prioritizing critical, actively exploited vulnerabilities.
$ GET /device/{token}/software?min_cvss_score=7.0&exploit=true
Reset a policy's pass/fail results and membership immediately — useful after remediating a widespread misconfiguration before the next scheduled check.
$ POST /api/v1/fleet/policies/42/reset
Adds ability to target policies to hosts using a combination of include and exclude labels.
Adds policy check before installing Windows and Linux setup experience software, skipping installs when the policy already passes to speed up end-user setup.
Changes calendar remediation events to be scheduled on the next business day (skipping weekends) after a policy failure.
Adds automation runs table to the policy details page showing per-host automation outcomes with filtering, search, and a reset policy action.
Adds per-host activity log entries when policy automations (webhook, tickets, Google Calendar, and Microsoft conditional access) fail or succeed.
+23 moreshow less
Adds POST /api/v1/fleet/policies/:policy_id/reset endpoint to reset a policy's pass/fail results, clearing counts and membership immediately.
Adds GET /api/v1/fleet/policies/:id/automation_activities endpoint to list automation activities for a policy.
Adds ability to keep Fleet-maintained apps automatically updated, pinned to a specific version or major version, or rolled back to a previously cached version, from the UI and via GitOpsGitOpsA practice where Git repositories serve as the single source of truth for infrastructure and application state, with automated systems continuously reconciling live environments to match what is declared in version control. (Fleet Premium).
Surfaces .sh script-only software packages on the macOS tab of Controls > Setup experience > Install software, tracked independently from the Linux tab.
Adds setup_experience_platform field on software packages in GitOps YAML for declarative per-platform setup experience selection.
Adds support for pre-install query, post-install script, and uninstall script on script-only packages (.sh and .ps1) via the UI, 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., and GitOps.
Adds error on the Windows enrollment status page (ESP) when setup experience software fails during automatic enrollment and Cancel setup if software fails is off.
Adds ' Support' as a new default self-service software category.
Adds support for $FLEET_VAR_HOST_* variables in Android configuration profiles and managed app configuration.
Adds automatic resending of Android certificate templates and managed app configurations when IdP variable values change.
Adds support for defining the default Fleet team BYO Apple devices enroll into.
Adds a Google Workspace integration that maps IdP users to hosts, populating IdP host vitals directly from your Google Workspace directory.
Adds an activity feed entry when a user runs a custom Apple or Windows MDMMDMMobile device management, centrally configuring and managing enrolled devices. Administrators use it to apply settings, distribute applications and enforce device security policies. command, visible in global and host activity feeds.
Adds an activity log entry when editing the managed local account setting via the update fleet endpoint or GitOps.
Enables tracking of mobile devices in the hosts online chart with default filtering that excludes mobile platforms.
Starts collecting non-critical CVEs, filtering them out of vulnerability charts by default.
Adds ability to filter vulnerable software by severity (CVSS score) and known exploit status on Fleet Desktop My device > Software tab (Fleet Premium), with min_cvss_score, max_cvss_score, and exploit query parameters added to GET /device/{token}/software.
Adds more filtering options and ability to set default filters for the Vulnerability Exposure chart via GitOps.
Adds support for all IdP variables and host platform in certificate template subject names and SANs.
Adds private network IP blocking to Fleet's HTTP client; loopback and cloud metadata addresses always blocked, RFC 1918RFC 1918IETF 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. ranges blocked by default, with --allow_private_network_integrations to allow them for on-prem integrations.
Adds s3.carves_cleanup_disabled server setting to skip S3 file carve reconciliation for deployments relying solely on bucket lifecycle policies.
Adds s3.carves_cleanup_max_per_run and s3.carves_cleanup_concurrency server settings to tune S3 carve cleanup throughput.
Adds configurable batch size FLEET_MDM_ANDROID_BATCH_SIZE (default: 1000 hosts) for Android MDM operations to prevent overwhelming the Google Android Management API.
28 details in Everything
└──▷ BREAKING ON UPGRADE
!RFC 1918 and other private network ranges are now blocked by default in Fleet's HTTP client; deployments with on-prem integrations (e.g. EJBCA, Jira, SCEPSCEPSimple Certificate Enrollment Protocol, an IETF-standardized protocol for automating X.509 certificate issuance and renewal over HTTP, letting cyber tools request PKI certificates from a CA without manual intervention. servers on private networks) must add --allow_private_network_integrations to restore connectivity.
Microsoft Defender is a built-in Windows security tool that protects devices from viruses, malware, ransomware, and other threats through real-time monitoring and scanning.
Microsoft launches Defender Experts Threat Intelligence service and expands MDRMDRManaged detection and response, an outsourced service that monitors security signals, investigates threats and helps respond to incidents. It combines specialist analysts with detection tools. to multi-cloud/third-party coverage via Sentinel.
New Microsoft Defender Experts Threat Intelligence service delivers curated, expert-interpreted threat intelligence tailored to your industry, geography, and environment — including early-warning campaign alerts, evolution updates, and recurring expert briefings.
Microsoft Defender Threat Intelligence (MDTI) capabilities are now fully converged into the Defender portal, making intelligence available in real-time across detection, investigation, response, hunting, and automation without context switching.
Microsoft Defender Experts MDR (formerly Defender Experts for XDRXDRExtended detection and response, combining signals and investigation across endpoints, networks, identities or cloud services. It helps analysts connect activity that separate tools might miss.) adds a new Plan 2 tier with third-party and multi-cloud coverage powered by Microsoft Sentinel, enabling expert-led monitoring and response across non-Microsoft cloud, identity, email, network, and endpoint sources.
Defender Experts MDR Plan 2 provides cross-platform threat analysis correlating Microsoft and non-Microsoft signals into a single incident narrative with vendor-aware guidance.
4 details in Everything
└──▷ BREAKING ON UPGRADE
!Microsoft Defender Experts for XDR is renamed to Microsoft Defender Experts MDR (existing capabilities carry forward as Plan 1).
CrowdStrike is a cloud-native cybersecurity platform that protects endpoints, cloud workloads, and identities from cyber threats using AI-powered threat detection and response.
CrowdStrike launches Falcon AIDR, a new runtime security category detecting and blocking threats targeting and originating from AI agents across endpoint, SaaSSaaSSoftware delivered over the internet and run on the provider's infrastructure, so the buyer pays a subscription instead of installing or operating anything locally., and cloud.
Introduces Falcon AI Detection and Response (AIDR), a new security category providing unified runtime visibility, detection, and automated response for AI agents across endpoints, SaaS, and cloud environments.
Discovers shadow AI by illuminating employee agent and tool usage across endpoints, AI gateways, 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. servers, cloud environments, and Copilot ecosystems.
Enforces granular attribute-based access controls aligned to AI governance policy, including policy based on user and model version.
Automatically detects and blocks confidential data, PIIPIIPersonally Identifiable Information — data that can identify a specific individual, such as names, email addresses, or SSNs. Cyber tools flag or redact it to limit exposure and meet privacy regulations., secrets, keys, and regulated data before exposure, with multiple redaction methods including format-preserving encryption.
Generates comprehensive audit trails of AI runtime behavior for compliance, forensics, and continuous improvement.
+1 moreshow less
Upcoming: deep Falcon sensor integration to discover and monitor AI agents (Claude Code, Cursor, Copilot, Kiro, Codex, shadow agents) at the OS/kernel level with a live fleet inventory, risk scoring, behavioral detection, and a unified agent execution chain + OS process tree causal graph for investigation and enforcement.
Trend Vision One is a unified cybersecurity platform that provides extended detection and response (XDR) across endpoints, networks, email, and cloud environments.
Trend Vision One adds Global approved list for teams, letting admins exempt specific collaboration spaces from ATP and DLPDLPData Loss Prevention: a category of controls that inspect and block sensitive data from leaving an organisation via email, web, or removable media, giving cyber tools a policy enforcement layer over data in motion and at rest. scanning.
└──▷ HOW TO FIND IT
Exempt a trusted internal team from ATP and DLP scanning to reduce noise and improve performance for that collaboration space.
📍In the console, go to Email and Collaboration Security → Cloud Email and Collaboration Protection → Policies, then open the Global approved list for teams and add the trusted team.
New Global approved list for teams lets administrators designate trusted teams to bypass Advanced Threat Protection and Data Loss Prevention inspection while keeping full protection elsewhere.
Reduces alert noise and improves performance by scoping security scanning away from known-safe collaboration spaces.
Trend Vision One lets admins suppress notifications for Pass-action detections with a single global toggle.
└──▷ HOW TO FIND IT
Reduce alert fatigue by silencing Pass-action notifications across your org — useful when high-volume benign detections are flooding admin and user inboxes.
📍In the console, go to Email and Collaboration Security → Cloud Email and Collaboration Protection → Policies and enable Skip notification for Pass actions.
Adds a global Skip notification for Pass actions setting that suppresses admin and end-user alerts when a detection results in no action taken on an email or file.
Tenzir is an open-source security data pipeline platform that enables teams to collect, route, transform, and analyze security telemetry at scale.
Tenzir v6.7.0 adds fork_merge for fanning out subpipelines and a force-stop action for pipelines.
New fork_merge operator fans out subpipelines over the same input and merges their outputs into a single stream.
Adds a force-stop action for pipelines in the Tenzir Platform.
Improved Tenzir Platform performance.
Restores unstructured (raw-log) ingestion support for Google SecOpsSecOpsThe practice of merging security and IT operations teams so that threat detection, response, and infrastructure management share the same workflows, giving cyber tools a single pipeline from alert to remediation..
Tenzir Platform v1.37.0 adds custom Git-backed package libraries, force-stop for pipelines, and a workspace key minting command.
Adds custom library sources: organizations can add public or private Git repositories alongside the built-in Tenzir Library to maintain their own package libraries.
Adds a Force stop action for pipelines that are slow to drain, enabling operators to hard-terminate stalled pipelines.
Tines is a no-code automation platform that enables security and operations teams to build powerful workflows by connecting tools and automating repetitive tasks.
Tines lets tenant owners restrict which users can create API keys via a new tenant-level setting.
└──▷ HOW TO FIND IT
Lock down API key creation to prevent non-owner users from generating credentials.
📍In the Tines console, go to Settings › API keys and enable the restriction option under the tenant settings section.
Adds DD_WATSON_SEARCH_ENABLED toggle to control Watson search indexing in DefectDojo.
└──▷ GET THIS VERSION
$ git clone --branch 3.1.101 https://github.com/DefectDojo/django-DefectDojo.git
# already have the repo? check out this version:$ git checkout 3.1.101
└──▷ USE IT
Disable Watson search indexing to reduce database load in environments where full-text search is not needed.
ini
DD_WATSON_SEARCH_ENABLED=False
New DD_WATSON_SEARCH_ENABLED setting in settings.dist.py / local_settings.py to gate Watson search indexing on or off.
Immuta is a data security platform that automates access control, privacy protection, and policy enforcement across cloud data environments.
Immuta adds automated Redshift data discovery, built-in managed groups, Delta Sharing row/column masking, Teradata hashing, and 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.
└──▷ HOW TO FIND IT
Use the embedded documentation chatbot to answer governance questions without leaving the Immuta UI.
📍In the Immuta UI, click the Documentation icon in the top menu, then type your question in natural language in the AI-powered documentation chatbot.
Introduces three built-in Immuta-managed groups — All Agents, All Users, and All Identities — available for use in subscription and data policies without manual membership management.
Enables row filtering and column masking (subscription and data policies) on Unity Catalog managed, external, and foreign tables in Databricks Delta Sharing shares.
Supports OAuth 2OAuth 2An 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..0 authentication for Immuta's Alation integration alongside existing API key authentication, eliminating recurring token rotation.
Extends Immuta's Databricks Unity Catalog tag ingestion to foreign catalogs (Lakehouse FederationLakehouse FederationA data architecture pattern that lets a single query engine read from multiple external data sources—databases, data lakes, warehouses—without moving or copying the data, reducing pipeline complexity for tools that need broad data access.) by falling back to system.information_schema tag views.
Adds constant and hash masking support for Immuta's Teradata integration, expanding beyond existing null masking.
+4 moreshow less
Slack and Microsoft Teams notifications for the Request app are now generally available, delivering real-time alerts for approvals, data access changes, and requests.
Extends Microsoft Teams Request app notifications to the Asia Pacific region (previously North America and Europe only).
Embeds Immuta documentation and an AI-powered documentation chatbot directly in the Immuta UI via a Documentation icon in the top menu.
Immuta's Starburst (Trino) integration now treats Trino principal names as case-insensitive when matching users and service accounts.
9 details in Everything
└──▷ BREAKING ON UPGRADE
!Beginning August 1, disabling a data source in Immuta will revoke data owners' Immuta-managed grants in the data platform for that data source; previously, disabling a data source did not remove those grants.
!Beginning August 1, Starburst (Trino) principal name matching becomes case-insensitive for all accounts; setups relying on case-sensitive principal name matching will be affected.
1 more release in this issue
· seen 2026-07-15 → 2026-07-16
Immuta adds automated Redshift discovery, built-in managed groups, OAuth for Alation, Delta Sharing masking, and Teams/Slack notifications.
└──▷ HOW TO FIND IT
Access Immuta documentation and ask governance questions in natural language without leaving the product.
📍In the Immuta UI, click the Documentation icon in the top menu to open the embedded docs or AI-powered chatbot, then type your question in natural language.
Introduces three built-in system-managed groups — All Agents, All Users, and All Identities — usable in both subscription and data policies without manual membership management.
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. authentication for Immuta's Alation integration, eliminating the need for recurring API key rotation.
Enables row filtering and column masking on Unity Catalog managed, external, and foreign tables in Databricks Delta Sharing shares (now GA for all accounts).
Adds Slack and Microsoft Teams notifications for the Request app (generally available), delivering event-driven alerts for approvals, data access changes, and requests.
Expands Microsoft Teams Request app notifications to the Asia Pacific region (public preview).
+4 moreshow less
Adds constant and hash masking support for Teradata in addition to existing null masking, improving cross-platform masking policy consistency.
Enhances Review Assist (AI-powered access request recommendation engine) to incorporate request form responses as contextual input for its analysis.
Embeds Immuta documentation and an AI-powered documentation chatbot directly in the Immuta UI via a Documentation icon in the top menu.
Starburst (Trino) principal name matching is now case-insensitive by default, enabling more reliable user and service account identification.
9 details in Everything
└──▷ BREAKING ON UPGRADE
!When a data source is disabled in Immuta, Immuta now revokes data owners' Immuta-managed grants in the data platform; previously, disabling a data source did not remove those grants (on by default 8/1, enabled for all 9/1).
Zscaler Data Protection is a cloud-native DLP solution that prevents sensitive data leakage across web, cloud applications, email, and endpoints.
Zscaler DLPDLPData Loss Prevention: a category of controls that inspect and block sensitive data from leaving an organisation via email, web, or removable media, giving cyber tools a policy enforcement layer over data in motion and at rest., DSPMDSPMData security posture management, locating sensitive data and checking how it is protected and who can reach it. It helps teams prioritize risks around their data stores., and inline proxy now detect and block SSLKEYLOGFILE TLS session-key material across endpoints, cloud storage, and network traffic.
Adds endpoint DLP detection for SSLKEYLOGFILE content matching legacy and newly standardized formats, including obfuscated or renamed files, to catch TLS key material exfiltration on user devices.
Extends DSPM scanning to identify files containing TLS session secrets across SaaSSaaSSoftware delivered over the internet and run on the provider's infrastructure, so the buyer pays a subscription instead of installing or operating anything locally. platforms, public cloud, and on-premises storage regardless of naming convention or file type.
Enables inline DLP engine to detect and policy-enforce outbound SSLKEYLOGFILE transfers in real time via file uploads, web channels, non-web channels, and email — with options to block, alert, or trigger investigation workflows.
Nudge Security discovers and manages SaaS applications across an organization, helping security teams gain visibility and govern employee-introduced tools through automated nudges.
Nudge Security adds two AI agents to automatically assess and remediate OAuth grant and browser extension risk.
New OAuth Grant Risk Analyst AI agent reviews newly discovered OAuth grants, flags overly permissive or high-risk ones, and resolves them by automatically revoking access or nudging the grantor — with human-in-the-loophuman-in-the-loopHuman in the loop, involving a person in an otherwise automated workflow. The person can review, approve, correct or supply information before work continues. oversight.
New Browser Extension Risk Analyst AI agent inspects installed browser extension artifacts and metadata, flags malicious, compromised, or misrepresented extensions, and orchestrates disable and uninstall workflows.
1Password is a password manager that securely stores, generates, and autofills passwords, credentials, and sensitive information across devices and browsers.
1Password for Linux adds a local 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 so AI assistants can access 1Password Developer tools.
└──▷ HOW TO FIND IT
Enable the local MCP server so an AI assistant (e.g., Claude, Cursor) can query 1Password Developer tools during a session.
📍In the 1Password app, go to Settings › Developer and start the local MCP server from the Developer tab.
New local MCP server in the Developer tab lets AI assistants connect to 1Password Developer tools.
Watchtower descriptions now explicitly surface scanning for plaintext secrets, including .env files and SSH keys.
2 details in Everything
3 more releases in this issue4 more releases in this issue · 2026-07-14 → 2026-07-15
1Password adds a local 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 for AI assistant integration and expands Watchtower plaintext secret scanningsecret scanningChecking files, code or other data for exposed credentials such as passwords, API keys and tokens. Findings help teams remove the secret and revoke or rotate it. visibility.
└──▷ HOW TO FIND IT
Connect an AI assistant to 1Password Developer tools by starting the local MCP server from the app.
📍In the 1Password app, go to Settings › Developer, then enable and start the local MCP server. Point your AI assistant's MCP client configuration to the local server endpoint exposed there.
New local MCP server in the Developer tab lets AI assistants access 1Password Developer tools.
Watchtower and Developer homepage descriptions now surface scanning coverage for plaintext secrets, including .env files and SSH keys.
1Password for Windows adds a local 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 for AI assistant access to Developer tools and expands Watchtower secret scanningsecret scanningChecking files, code or other data for exposed credentials such as passwords, API keys and tokens. Findings help teams remove the secret and revoke or rotate it. descriptions.
└──▷ HOW TO FIND IT
Connect an AI assistant (e.g., Claude, Cursor) to your 1Password vault by starting the local MCP server from the Developer tab.
📍In the 1Password app, go to Settings › Developer, then enable and start the local MCP server to expose 1Password Developer tools to your AI assistant.
Adds a local MCP server launchable from the Developer tab, enabling AI assistants to access 1Password Developer tools.
Watchtower descriptions on the Developer homepage now explicitly mention scanning for plaintext secrets, including .env files and SSH keys.
1Password in the browser adds toggleable extension diagnostic logging in Settings > Help.
└──▷ HOW TO FIND IT
Enable diagnostic logging when troubleshooting extension behavior to capture logs for 1Password Support.
📍In the browser extension, go to Settings › Help › Share diagnostic logs with 1Password and toggle logging on.
Adds a toggle to enable or disable extension diagnostic logging via Settings > Help > Share diagnostic logs with 1Password.
Improves SSOSSOSingle sign-on, using one identity provider to sign in to multiple applications. It reduces separate passwords and centralizes account and access management. account migration so the desktop app only comes to the front when the migrating account is already configured in the desktop app.
iDenfy is an identity verification platform that automates KYC/AML compliance through AI-powered document scanning, facial recognition, and liveness detection for businesses.
iDenfy adds Brazil Digital CNH, new KYBKYBKnow Your Business — a regulatory compliance process for verifying the legal identity, ownership, and legitimacy of a business entity, used by cyber tools to screen counterparties or customers against fraud and sanctions risk. automation API, Risk Assessment PDF export, Arabic/RTLRTLRegister Transfer Level, a hardware design abstraction describing how data moves between registers each clock cycle; cyber tools use it to analyze or synthesize digital circuits from HDL source. IDV support, and multi-theme KYCKYCKnow Your Customer — a regulatory requirement that financial services verify the identity of their users before granting access, used by cyber tools handling payments or financial data to stay compliant. branding.
Adds Brazil Digital CNH support for identity verification.
New API endpoint to re-run KYB automation workflows.
Dedicated tax identification fields (VAT Number, EIN, TIN) added to KYB workflows.
External Reference field now available in KYC verification reports.
Risk Assessment PDF export now supported.
+30 moreshow less
Dashboard records can be opened in a new browser tab.
New digital ID support for India and the Philippines.
PAYG social login now available.
Configurable KYB company expiration period.
KYB Verify Later automated IDV reminders.
AI reviewer spelling variation matching for KYB.
Arabic language and RTL support for IDV (Sprint 127).
KYC multi-theme branding support (Sprint 127).
KYB sole proprietorship identity verification (Sprint 127).
Dynamic workflow decline action (Sprint 127).
New Digital ID support: Austrian Handy Signatur, Freja eID, Mobile ID (Sprint 126).
Auto document type detection for direct processing (Sprint 126).
Companies House integration for KYB (Sprint 125).
Custom redirect URL for iFrame (Sprint 125).
KYB Dynamic Workflows (Sprint 124).
Face capture guidance added to IDV flow (Sprint 124).
Korean and Traditional Chinese language support added (Sprint 124).
SMSSMSShort Message Service, the protocol built into cellular networks for sending text messages between phones; cyber tools use it for alerting, two-factor authentication delivery, or phishing simulation over mobile channels. verification support added (Sprint 123).
AI KYB Reviewer fraud prevention capabilities (Sprint 122).
SMS expansion for broader geographic coverage (Sprint 122).
Watermarks on exported PDFs (Sprint 122).
Android SDK: 3D liveness updates and new eID support.
Android SDK: Arabic locale with RTL support and redesigned bottom sheets.
Android SDK 8.7.6: Reduced SDK size and optional 3D liveness.
Android SDK 8.7.5: Japanese locale support and new alert dialogs.
Android SDK 8.7.1: Screenshot blocking and virtual camera detection.
iOS SDK: 3D liveness updates and new eID support.
iOS SDK: Arabic locale with RTL support.
iOS SDK 8.7.7: NFCNFCNear Field Communication, a short-range (≤4 cm) wireless standard maintained by the NFC Forum, enabling data exchange between devices; cyber tools use it to probe, emulate, or intercept contactless card and tag transactions. improvements.
iOS SDK 8.7.6: KYC questionnaire multilanguage support and EID document support.
Clerk is a complete user authentication and management platform that provides embeddable UI components, APIs, and dashboards for adding secure sign-in, sign-up, and user management to applications.
Clerk 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. connections can now send an email address, custom attribute, or no login hint to your IdPIdPA service that authenticates users and supplies identity information to other applications. It is the sign-in authority behind many single sign-on systems..
Supports per-SAML-connection configuration of the login hint sent to identity providersidentity providersA service that authenticates users and supplies identity information to other applications. It is the sign-in authority behind many single sign-on systems.: email address, a custom attribute, or none.
Zscaler is a cloud-native security platform that provides secure internet access, zero trust networking, and threat protection for users, devices, and applications.
Zscaler launches Autonomous Application Shield to auto-apply app-specific protections in the patch gap via App Connectors and global threat intel.
└──▷ HOW TO FIND IT
Request early access to Autonomous Application Shield to get hands-on experience and direct roadmap input before general availability.
📍In the Zscaler console, contact your Zscaler representative or register via the Early Access Program webinar link to enable Autonomous Application Shield for your private applications.
New Autonomous Application Shield capability continuously assesses private application exposure via App Connectors and automatically applies per-app protections in the Zscaler cloud without requiring a patch.
Per-application protection reasoning: the Zscaler cloud evaluates each private application individually and applies only the controls that specific application needs, reducing false positives and performance impact.
Automatic protection adaptation when developers ship new releases — security policy updates without manual re-tuning, policy review meetings, or security team intervention.
Global threat intelligence integration feeds zero-dayzero-dayA vulnerability for which defenders have had no prior opportunity to apply a fix, often because it was previously unknown. A zero-day exploit takes advantage of that gap. and emerging attack technique data from Zscaler's worldwide customer base into the protection engine, with ML continuously tuning policies per application.
DNSFilter is a cloud-based DNS security platform that protects users from malware, phishing, and unwanted content by filtering DNS queries in real time.
DNSFilter Identities framework adds direct Entra IDEntra IDMicrosoft's cloud-based identity and access management service, formerly Azure Active Directory, used by cyber tools to authenticate users and enforce access policies in Microsoft 365 and Azure environments./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. integration, rebuilt Collections, and MSPMSPA managed service provider, a company that operates IT services for customers. It may manage systems, networks, backups or other ongoing technology needs.-level identity management.
└──▷ HOW TO FIND IT
MSPsMSPsA managed service provider, a company that operates IT services for customers. It may manage systems, networks, backups or other ongoing technology needs. managing multiple organizations can configure synced policy assignment for all orgs from one place instead of entering each org individually.
📍From the MSP dashboard, navigate to Identities to create or manage Collections and Identity Connections across all organizations in a single view.
New Identities framework in the DNSFilter dashboard unifies users, collections, and policy assignment across identity sources for enterprise and MSP environments.
Direct Microsoft Entra ID integration via SCIM provisioning eliminates the AD Sync Tool dependency, with support for Entra Hybrid environments.
Rebuilt Collections engine supports large directories, multiple identity sources, and high-complexity group structures.
New unified Identities view gives administrators a consolidated picture of all identity sources, user objects, collections, and policy assignments.
MSP dashboard now supports managing Users, Collections, and Identity Connections across all organizations from a single view, without navigating into each organization individually.
Zscaler Internet Access is a cloud-based secure web gateway that protects users by inspecting and filtering internet traffic to block threats and enforce policies.
Zscaler Endpoint Context enriches security logs with endpoint data 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. and SOCSOCA security operations center, the team or function that monitors threats, investigates alerts and coordinates incident response. Its work can be internal, outsourced or shared. automation workflows.
Zscaler Private Access is a zero trust network access solution that securely connects users to private applications without exposing them to the network.
Zscaler adds Endpoint Context policy enforcement and Autonomous Application Shield for continuous private-app protection.
New Endpoint Context capability feeds endpoint-derived intelligence into Zero Trust Exchange policy decisions across TLS/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. inspection, Zero Trust Firewall, DNS security, intrusion prevention, and Advanced Threat Protection.
New Autonomous Application Shield continuously assesses private application exposure via App Connectors and automatically applies app-specific protections in the Zscaler cloud to cover the window between vulnerability disclosure and patching.
RegScale is a continuous compliance automation platform that helps organizations manage security controls, audits, and regulatory requirements in real time.
RegScale 6.31.1.5 adds custom rules support for POA&MPOA&MA Plan of Action and Milestones: a US federal government document that tracks known security weaknesses in a system, the remediation steps planned, and the deadlines for completing them, required under FISMA compliance. configuration.
$ git clone --branch rust-v0.145.0-alpha.16 https://github.com/openai/codex.git
# already have the repo? check out this version:$ git checkout rust-v0.145.0-alpha.16
Tracks prompt cache write token usage (cache_write_input_tokens) and reports it in analytics, metrics, traces, and rollout data alongside existing cache-read usage.
Strengthens forced rm command detection to catch forced removals inside control flow, substitutions, pipelines, traps, nested shells, and sudo/env wrappers — including combined, separate, and long-form force options even when they follow operands.
Extends image generation timeout in code mode to 120 seconds for both the initial call and subsequent waits.
Improves turn history summaries to use only final_answer-phase agent messages, excluding commentary items.
4 details in Everything
└──▷ BREAKING ON UPGRADE
!The template_id field is removed from 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. tool call items, lifecycle events, and McpToolCallAppContext app-server responses — any client or integration reading template_id (or templateId) from these payloads will no longer find it.
0 more releases in this issue2 more releases in this issue · 2026-07-15 → 2026-07-16
Codex CLI alpha.15 adds external agent memory migration, paginated subagent history, and smarter 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. capability-root propagation.
└──▷ GET THIS VERSION
$ git clone --branch rust-v0.145.0-alpha.15 https://github.com/openai/codex.git
# already have the repo? check out this version:$ git checkout rust-v0.145.0-alpha.15
Adds external agent memory migration: discovers project memory Markdown files, preserves project scope, copies selected projects into the Codex memory extension workspace, and enqueues memory consolidation after workspace changes.
Propagates deferred environment capability roots to MCP, making executor plugins available as soon as their environment is ready and refreshing the MCP runtime when the selected root set changes.
Preserves paginated history for spawned or forked subagents, inheriting the parent's model context as a rollout prefix and resuming from stored model context instead of legacy rollout history.
Warns when conflicting capability root locations are registered under the same ID, retaining the first location and logging the conflict instead of silently dropping it.
Selects the elevated Windows sandbox automatically when network proxies are in use.
Codex CLI adds Frameless Bidi realtime voice, 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. cache opt-out, Cursor import support, and paginated thread history.
└──▷ GET THIS VERSION
$ git clone --branch rust-v0.145.0-alpha.14 https://github.com/openai/codex.git
# already have the repo? check out this version:$ git checkout rust-v0.145.0-alpha.14
Adds realtime conversation protocol v3 with Frameless Bidi support, translating audio, transcripts, handoffs, session context, and lifecycle events over WebSocketWebSocketA protocol that upgrades an HTTP connection into a persistent, full-duplex channel, letting a server push data to a client without repeated polling. Cyber tools use it for real-time event streaming and live command output. and WebRTCWebRTCA browser-native, open standard for real-time peer-to-peer audio, video, and data exchange without plugins. Cyber tools use it to tunnel traffic or exfiltrate data through connections browsers treat as legitimate..
Allows MCP servers to opt out of tool catalog caching via the experimental codex/tool-catalog-cache capability (cacheable: false), preventing stale snapshots from being served.
Adds Cursor IDE support to the setup import flow, joining existing editor integrations.
Migrates plugin commands into skills automatically on install, converting legacy plugin command definitions to the Codex skill format.
Scans skill roots concurrently (up to eight at a time) for faster startup on repos with many skills.
Autonomous coding agent as an SDK, IDE extension, or CLI assistant.
Cline CLI v3.0.41 adds compaction progress 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., model ID suggestions from OpenAI-compatible endpoints, and persistent workspace git info.
└──▷ GET THIS VERSION
$ git clone --branch cli-v3.0.41 https://github.com/cline/cline.git
# already have the repo? check out this version:$ git checkout cli-v3.0.41
Shows compaction progress status in the TUI so you can track context-reduction operations in real time.
Suggests model IDs from OpenAI-compatible endpoints when configuring a provider, reducing manual lookup.
Persists and refreshes workspace git branch and remote info across sessions.
Blocks compaction from running during an active turn, preventing interference with in-progress responses.
Cline SDK v0.0.61 adds context compactioncontext compactionCondensing an AI conversation or working history to free space in the model's context window. A summary preserves useful information, but details can be lost. progress, Plan/Act system-prompt guidance, and OpenAI-compatible model ID suggestions.
└──▷ GET THIS VERSION
$ git clone --branch sdk/sdk/v0.0.61 https://github.com/cline/cline.git
# already have the repo? check out this version:$ git checkout sdk/sdk/v0.0.61
Adds progress status reporting while context compaction runs, giving visibility into long-running compaction operations.
Persists and refreshes workspace git info (branch/remote) across sessions so agents maintain accurate repo context.
Adds Plan/Act mode guidance to the system prompt, with nudges when switching between modes.
Suggests model IDs from OpenAI-compatible endpoints when configuring custom providers.
Centralizes provider-specific request headers in the 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. layer for consistent multi-provider header handling.
+2 moreshow less
Attaches organization context to telemetry when identifying with cached credentials.
Introduces a shared @cline/ui theme package for consistent UI theming across Cline surfaces.
Pi v0.80.8 adds unified model/auth runtime, live catalog refresh, and xAI Grok 4.5 with device-code OAuth.
└──▷ GET THIS VERSION
$ git clone --branch v0.80.8 https://github.com/earendil-works/pi.git
# already have the repo? check out this version:$ git checkout v0.80.8
└──▷ TRY IT
Force a full model catalog refresh after adding a new provider, without reinstalling pi or extensions.
$ pi update --models
Adds ModelRuntime as the canonical async SDK facade, centralizing model configuration, provider authentication, and dynamic provider catalogs.
Adds pi update --models command to force an immediate model catalog refresh without updating pi or extensions.
Adds /model live background refresh of configured providers, updating the selector with partial results as they arrive.
Adds provider-owned /login discovery, surfacing ambient auth status and informational links directly from registered providers.
Adds xAI device-code OAuth login so users can authenticate to xAI interactively without an API key.
+3 moreshow less
Adds Grok 4.5 OpenAI Responses support with selectable low, medium, and high thinking levels.
Adds file-backed dynamic model catalogs (models-store.json), per-provider pi.dev catalog overlays, and Radius gateway support.
Adds refreshModels(context) extension hook for dynamic model discovery with optional provider-controlled persistence.
8 details in Everything
└──▷ BREAKING ON UPGRADE
!The CreateAgentSessionOptions.authStorage and modelRegistry SDK options are replaced by the async modelRuntime option; AuthStorage and its storage backends are no longer exported — use ModelRuntime, a custom pi-ai CredentialStore, or readStoredCredential() instead.
!ModelRuntime.getAll(), find(), getSnapshot(), and getAuthOptions() are removed; use pi-ai Models methods getModels(), getModel(), getProviders(), and checkAuth() directly.
!Request-auth assembly via ModelRegistry.getApiKeyAndHeaders() is replaced by ModelRuntime.getAuth(); passing a provider ID returns provider-scoped auth, passing a model also resolves built-in, models.json, and extension model headers.
!ModelRegistry.refresh() changed from synchronous void to Promise<void>; extensions must await it before making synchronous registry reads.
Crush v0.85.0 adds an interactive question tool, an elapsed-time timer, and LM Studio vision support.
└──▷ GET THIS VERSION
$ git clone --branch v0.85.0 https://github.com/charmbracelet/crush.git
# already have the repo? check out this version:$ git checkout v0.85.0
Adds a first-class question tool letting LLMsLLMsA 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. prompt users with single-choice, multiple-choice, free-form text, and yes/no forms — with full mouse support.
Displays an elapsed-seconds timer so you can see how long the 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. has been running.
Enables LM Studio vision capabilities (image support) via SupportsImages.
Adds automatic transparent auth retry so dropped credentials are refreshed without user intervention.
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 v2.1.211 adds subagent stream visibility, cross-worktree permission persistence, and scientific-notation env vars.
└──▷ GET THIS VERSION
$ git clone --branch v2.1.211 https://github.com/anthropics/claude-code.git
# already have the repo? check out this version:$ git checkout v2.1.211
└──▷ TRY IT
Capture subagent reasoning in a CI pipeline that parses stream-json to audit what sub-tasks were run.
$ claude --output-format stream-json --forward-subagent-text-p 'Refactor the auth module and write tests' | tee agent-stream.ndjson
Override a token-budget environment variable using scientific notation — useful in shell configs where large numbers are error-prone.
$ CLAUDE_CODE_MAX_OUTPUT_TOKENS=1e6 claude -p 'Summarize the entire repo'
Adds --forward-subagent-text flag and CLAUDE_CODE_FORWARD_SUBAGENT_TEXT env var to surface subagent text and thinking inside --output-format stream-json output.
Changes always allow permission rules to save at the repository root, so approvals granted in one git worktree persist across all worktrees and sessions.
Accepts scientific notation and digit-separator spellings (e.g. 1e6, 64_000) for integer environment variables such as timeouts, token budgets, and retry counts.
Enables save_to_disk on Claude-in-Chrome screenshot actions to actually write the image to disk and return its path.
Improves background agent result reporting so Claude waits for real completion and reports still-running agents instead of fabricating results.
+3 moreshow less
Adds Vim NORMAL-mode support for s (substitute char) and S (substitute line) in the editor input.
Improves the memory index over-limit warning to measure only loaded content, excluding frontmatter and HTML comments.
Changes /usage-credits to require confirmation before sending a request to organization admins.
Install and Run Python Applications in Isolated Environments
pipx 1.15.1 adds outdated, exec, reset, PEP 751PEP 751A draft Python packaging standard that defines a lock file format for recording exact dependency versions, giving cyber tools a reproducible, machine-readable way to pin their dependency trees. lock file support, scriptable JSON output, and major performance improvements.
└──▷ GET THIS VERSION
$ git clone --branch 1.15.1 https://github.com/pypa/pipx.git
# already have the repo? check out this version:$ git checkout 1.15.1
└──▷ TRY IT
Check which of your globally installed tools have newer versions available before deciding what to upgrade.
$ pipx outdated
Filter pipx list output to inspect a single installed package without scrolling through everything.
$ pipx list black
Start fresh after a corrupted pipx state — wipes all managed environments and reinitialises pipx.
$ pipx reset
New pipx outdated command lists all installed tools that have upgrades available.
New pipx exec command runs an already-installed application directly without re-resolving environments.
New pipx reset command wipes and rebuilds pipx's state from a clean install.
New pipx cache subcommand manages cached pipx run environments.
Adds PEP 751 lock file support for pipx install, enabling reproducible installs from lock files.
+17 moreshow less
Adds scriptable/machine-readable output format selection across install, inject, pin, uninstall, and other commands.
Adds --skip-maintenance flag (feat: allow skipping maintenance) to bypass shared-library upkeep during operations.
New pipx health repair mode automatically fixes broken virtual environments.
Enables pipx expose to selectively link dependency apps and completion scripts a package ships.
Supports pipx inject with main-package extras, allowing extras on the primary package to be activated at inject time.
Adds pipx install guidance for library-only packages that ship no CLI entrypoints.
Adds Python interpreter argument passthrough for pipx run.
Adds pipx list filtering by package name to show details for a single installed tool.
Adds pipx install support for persisting PEP 723PEP 723A Python Enhancement Proposal that standardizes embedding dependency metadata directly in single-file scripts, letting tools automatically resolve and install requirements without a separate config file. inline-scripted Python scripts as permanent tools.
Adds download progress bar when pipx install fetches packages.
Streams verbose installer output live during pipx install --verbose.
Installs a pipx man page from wheels.
Parallelizes update checks in pipx upgrade-all for significantly faster operation.
Multiple startup-path performance improvements (lazy venv scans, cached site-packages paths, deferred Python resolution) that speed up every pipx invocation.
Adds pipx manifest sync to reconcile a set of managed tools from a declarative manifest.
THE CYBER TOOLCHAIN