PAGES
Subscribe Get it delivered — five to ten releases we picked each morning from the tools you run, with every release in the issue on the web. Compare plans →

The Cyber Toolchain — issue 024, July 24, 2026

THE CYBER TOOLCHAIN NO. 024
Tail
THE DAILY RELEASE FIREHOSE
PUBLISHED JULY 24, 2026 · EVERY DAY
EDITIONS tail grep head ls diff uniq

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.

VIEW
ISSUE VIEW scroll
ISSUE WIDTH reading width
Do you prefer scroll?
ADDITIONAL FILTERS
$ tct list   # 25 tools matched
Application Security
◆  SAST / SCA / Code Security

Promptfoo

Sources Release page → updated · Jul 24, 2026 Notes

Promptfoo is an open-source tool for testing, evaluating, and red-teaming LLM applications to detect vulnerabilities, improve reliability, and compare model outputs.

└──▷ TRY IT
View promptfoo log output directly from the CLI for debugging rate limiting or provider issues.
$ promptfoo logs
  • Adds xAI Voice provider overrides: apiBaseUrl, websocketUrl, and function call support.
  • Adds test-level prompts filter, providers filter, and 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 individual test case level.
  • Adds --share and --no-share flags to Model Audit for controlling cloud sharing.
  • Adds multi-input variable support to red team scans for testing systems with complex input structures.
  • Adds early scan stop when plugins fail to generate test cases.
+1 more show less
  • Propagates -j flag to Python worker pools for Python concurrency in Enterprise.
Was this useful?

Portkey

Sources Release page → updated · Jul 24, 2026 Notes

Portkey is an AI gateway and observability platform that helps developers manage, monitor, and optimize LLM API calls with routing, caching, and reliability features.

Portkey adds vault-backed Secret References, weekly rate limits, cross-cloud GCS log delivery, and four new guardrail integrations.

└──▷ USE IT
Apply Zscaler Detection Policies to all 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. traffic through Portkey so your existing Zscaler policy class governs both inputs and outputs.
json
Add Zscaler AI Guard as a guardrail with your policyId:

{
  "guardrails": {
    "beforeRequestHook": {
      "provider": "zscaler",
      "policyId": "<your-zscaler-policy-id>",
      "timeout": 10000
    },
    "afterRequestHook": {
      "provider": "zscaler",
      "policyId": "<your-zscaler-policy-id>",
      "timeout": 10000
    }
  }
}
  • Adds Secret References so virtual keys and integrations can pull credentials at runtime from AWS Secrets Manager, Azure Key Vault, or HashiCorp Vault via secret_mappings, keeping secrets out of Portkey config.
  • Supports weekly budget and usage windows (rpw) for rate-limit policies so spend caps align with weekly planning cycles.
  • 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 cap.
  • Enables GCS log storage from AWS-hosted gateways via GCP Workload Identity Federation (GCP_WIF_AUDIENCE, GCP_WIF_SERVICE_ACCOUNT_EMAIL), eliminating long-lived cross-cloud credentials.
  • Lets organization admins include archived workspaces in analytics graphs and filters, with workspace slugs now accepted alongside IDs in saved filters.
+11 more show less
  • Integrates Zscaler AI Guard as a guardrail partner, applying Zscaler Detection Policies to LLM inputs and outputs via beforeRequestHook/afterRequestHook with a required policyId.
  • Integrates Akto Agentic Security as a guardrail partner to scan LLM traffic 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.
  • Adds customHost support for the Bedrock Guardrails plugin so evaluations route to private or regional Bedrock-compatible endpoints.
  • Adds required metadata key–value guardrails that block requests missing specified fields before they reach the model.
  • Adds DeepInfra provider support with 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), chat completions, and embeddings.
  • Extends DeepSeek support: deepseek-chat gains tools, tool_choice, and stream_options; deepseek-reasoner maps reasoning_effort to thinking mode and surfaces reasoning_content in streams.
  • Extends Bedrock batch inference to cover embeddings in addition to chat completions.
  • Maps Portkey metadata to Vertex AI resource labels and adds enterprise web search grounding via enterpriseWebSearch/enterprise_web_search.
  • Enables AWS workloads to reach Vertex AI through AWS–GCP Workload Identity Federation (GCP_WIF_AUDIENCE, GCP_WIF_SERVICE_ACCOUNT_EMAIL).
  • Adds Azure AI Foundry rerank support for Cohere rerank models (e.g. cohere.Cohere-rerank-v4.0-pro), with the gateway stripping the cohere. prefix for the provider.
  • Updates OpenTelemetryOpenTelemetryAn open standard and set of tools for collecting traces, metrics and logs from applications. It lets services export observability data without tying instrumentation to one monitoring vendor. GenAI semantic spans to semconv 1.40.0, adds OTELOTELAn open standard and set of tools for collecting traces, metrics and logs from applications. It lets services export observability data without tying instrumentation to one monitoring vendor. exporter support for guardrail flows, and supports custom resource attributes for downstream APMAPMApplication Performance Monitoring — a category of software that collects metrics, traces, and logs from running applications so developers can detect slowdowns and diagnose their root cause. and tracing.

16 details in Everything

Was this useful?

Veracode

Sources Blog / feed → SCA updates (2026-07-24) Blog

Veracode is a cloud-based application security platform that helps developers identify and remediate vulnerabilities in their code through static, dynamic, and software composition analysis.

Veracode 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. gains malicious package detection, Poetry support, 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. generation, new agent flags, and Unified Policy for workspaces.

└──▷ TRY IT
Capture structured JSON scan output for automation while keeping a readable summary visible in the terminal during a CI run.
$ srcclr scan . --json=results.json --show-cli
Assign a Unified Policy to an SCA workspace so builds break based on policy criteria — navigate to the workspace policy settings in the Veracode Platform.
📍In the Veracode Platform, go to SCA › Workspaces › <your workspace> › Policies, then select Assign Policy and choose or set a default Unified Policy.
  • Adds --appname flag to the SCA agent scan command (Commercial region) to link scan results directly to an application profile, requiring OAuth or HMACHMACA keyed hash that proves a message came from someone holding a shared secret and was not altered. Webhooks and API request signing commonly use it. authentication.
  • Adds --show-cli flag to the SCA agent scan command: used with --json, it saves structured JSON output to a file while also displaying a human-readable summary in the console.
  • Enables Unified Policy assignment to SCA agent-based scan workspaces in the US Federal region, allowing policies to create issues and break builds based on defined criteria.
  • Enables Unified Policy assignment to SCA agent-based scan workspaces in the European region.
  • Adds ability to configure the Vulnerability Severity policy rule (Commercial region) to break builds only when an upgrade path to fix a vulnerable component is available.
+14 more show less
  • Adds malicious package detection to SCA upload and agent-based scans — no configuration or agent update required to enable.
  • Adds support for generating SBOM files during SCA agent-based scans.
  • Adds support for scanning SBOM files as part of SCA upload scans.
  • Adds support for scanning SBOM files as part of SCA agent-based scans.
  • Adds support for Python's Poetry package manager in SCA Upload scans.
  • Adds support for Python's Poetry package manager in SCA agent-based scans.
  • Adds support for Python 3.12 environments in SCA agent-based scans.
  • Adds autopackaging support to the SCA agent (v3.8.114+), enabling automated packaging workflows without affecting existing scans.
  • Adds deep link support for SCA pages: unauthenticated users are redirected to login and then back to the original scan results link.
  • Adds vulnerability details side pane on the Vulnerabilities tab of the new SCA homepage (Beta).
  • Updated digital certificate for the SCA agent enables customers to verify source and integrity of the agent file before execution.
  • Changes default SCA agent download URL to https://sca-downloads.veracode.com (legacy https://download.sourceclear.com available for 12 more months).
  • Changes default backend API URL for Commercial region customers to https://sca-api.veracode.com (legacy https://api.sourceclear.io available for 12 more months).
  • Changes default backend API URLs for European and US Federal region customers to https://sca-api.veracode.eu and https://sca-api.veracode.us respectively.

19 details in Everything

└──▷ BREAKING ON UPGRADE
  • !The default SCA agent download URL changes from https://download.sourceclear.com to https://sca-downloads.veracode.com; the legacy URL will be retired after 12 months.
  • !The default backend API URL for Commercial region customers changes from https://api.sourceclear.io to https://sca-api.veracode.com; the legacy URL will be retired after 12 months.
  • !The default backend API URLs for European region customers change from https://platform-backend.analysiscenter.veracode.eu to https://sca-api.veracode.eu; legacy URLs will be retired after 12 months.
  • !The default backend API URLs for US Federal region customers change from https://platform-backend.analysiscenter.veracode.us to https://sca-api.veracode.us; legacy URLs will be retired after 12 months.
Was this useful?

Cloudsmith

Sources Blog / feed → CLI v1.19.0: Authenticate automatically with OIDC and Docker credential discovery (2026-07-23) Blog

Cloudsmith is a cloud-native, universal package management platform that enables teams to securely host, manage, and distribute software artifacts across multiple formats.

Cloudsmith CLI v1.19.0 adds automatic 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. credential discovery for CI/CD and a Docker credential helper for registry auth.

  • Automatically discovers OpenID ConnectOpenID ConnectOpenID 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. (OIDC) credentials for all major CI/CD platforms, eliminating the need for static API keys in pipelines.
  • New Docker credential helper automatically authenticates to Cloudsmith registries without manual docker login steps.

2 details in Everything

Was this useful?
◆  Secrets Detection & Management

GitGuardian Self-Hosted

Sources Blog / feed → 2026.7 - Required (2026-07-24) Blog

GitGuardian Self-Hosted is an on-premise security tool that detects and prevents secrets leakage in Git repositories.

GitGuardian Self-Hosted 2026.7 adds Public Monitoring, native Gerrit scanning, Smart Notifiers, Gateway API routing, and 9 new secret detectors.

└──▷ USE IT
Enable the Public Monitoring worker in your Helm values to start processing public perimeter secret-leak incidents.
yaml
celeryWorkers:
  public-monitoring:
    enabled: true
Deploy the GitGuardian 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 in your cluster (beta) by enabling it in your Helm values.
yaml
mcpServer:
  enabled: true
  • Adds Public Monitoring on self-hosted instances (license-gated): monitor your public perimeter for secret leaks and route public incidents through custom webhooks, Slack, Microsoft Teams, ServiceNow ticketing, and severity-filtered email alerts via the new celeryWorkers.public-monitoring worker.
  • Adds native Gerrit VCSVCSVersion control system: software that tracks changes to files over time, letting teams collaborate and roll back mistakes. Cyber tools use it to manage source code, configs, and signatures. source with full historical scanning, real-time detection via webhook plugin, and read-replica support for cloning.
  • Adds Smart Notifiers: define per-channel rules to filter alerts, tickets, and emails by severity, validity, secret type, or GitGuardian tag across internal and public monitoring.
  • GitHub check runs now scan oversized pull requests by cloning the repository branch instead of calling the GitHub API commit-by-commit, and can skip merge commits to protect shared API rate limits; thresholds and behavior are configurable from the admin area GitHub preferences.
  • Adds Gateway API routing to direct external traffic through the Kubernetes Gateway API instead of the legacy Ingress resource, plus support for HPA or KEDA autoscaling for webapps.
+5 more show less
  • Secrets Detection Engine v2.166 adds 9 new detectors: Amadeus OAuth Credentials, Kinde OAuth Credentials, TomTom API Key, Datadog Personal Access Token, and five Azure services (Maps, Mixed Reality, Service Bus, Quantum, Bot Direct Line), plus 2 new checkers (Gemfury Deploy/Push Token, Gemfury Full Access Token).
  • Jira Notifier now exposes remediation guidance and detector-specific links as assignable fields in the ticket description template.
  • Custom Sources (BYOS) gain an option to scan with validated detectors only.
  • The daily analytics job can now write Spark scratch data to a dedicated ephemeral volume instead of the node's local disk, preventing failures on large workspaces.
  • Adds opt-in mcpServer deployment (Helm only, beta) to run the GitGuardian MCP server in-cluster; disabled by default.

10 details in Everything

└──▷ BREAKING ON UPGRADE
  • !Helm installations: logCollector.collectMigrationLogs is now enabled by default and injects a native sidecar requiring Kubernetes 1.29 or later; on older clusters you must set logCollector.collectMigrationLogs: false.
  • !Endpoint Protection (Developer Endpoint Protection) is not functional on self-hosted instances in this release — do not upgrade if you rely on Endpoints without contacting GitGuardian support first.
Was this useful?
Cloud & Container Security
◆  Cloud Security Posture

Prowler

Sources Release notes →Source code → 5.36.0 Notes Code

Prowler is the world’s most widely used open-source cloud security platform that automates security and compliance across any cloud environment.

Prowler 5.36.0 adds SageMaker secrets detection, grouped Jira dispatch, cleaner Attack Paths UI, and regionless OCIOCIOpen Container Initiative standard, maintained by the Linux Foundation, defining the image format and runtime spec that container tools share so images built once run anywhere OCI-compliant. onboarding.

└──▷ GET THIS VERSION
$ git clone --branch 5.36.0 https://github.com/prowler-cloud/prowler.git
# already have the repo? check out this version:
$ git checkout 5.36.0
  • New sagemaker_notebook_instance_no_secrets check scans SageMaker notebook instance OnCreate and OnStart lifecycle scripts for hardcoded API keys, passwords, tokens, and connection strings.
  • Grouped Jira workflows (Prowler Cloud/Private Cloud): send selected findings or Finding Groups to Jira as one grouped issue or separate issues, with deep links and filter context preserved.
  • Attack Paths query selector now hides confirmed-empty queries for the selected scan, surfacing only paths that returned data; full catalog published on Prowler Hub.
  • OCI provider setup no longer requires a region — SDK handles region discovery automatically, simplifying onboarding.
  • New AI agents tutorials and overview banner guide users through connecting AI agents to Prowler Cloud via the API to query and act on security posture programmatically.
+1 more show less
  • Compliance overview ingest now runs in a single transaction per scan with a configurable DJANGO_COMPLIANCE_COPY_BATCH_SIZE (default 2000), reducing database write pressure at scale.

6 details in Everything

Was this useful?
◆  Runtime Security

Docker Sandboxes

Sources Release notes → v0.37.0 Notes

Docker Sandboxes isolates AI model execution environments to safely test and run untrusted code with controlled resource access.

Docker Sandboxes v0.37.0 adds SSH access to sandboxes, shared agent skills, port publishing, and richer audit records.

└──▷ GET THIS VERSION
$ git clone --branch v0.37.0 https://github.com/docker/sbx-releases.git
# already have the repo? check out this version:
$ git checkout v0.37.0
└──▷ TRY IT
Set up SSH access once so every sandbox becomes reachable by name from any OpenSSH-compatible client or remote-dev tool.
$ sbx setup ssh
Connect to a running sandbox for an interactive shell or one-shot command without leaving your terminal.
$ ssh my-sandbox.sbx
Import host agent skills into the shared store (preview first, then apply) so new sandboxes automatically inherit them.
$ sbx skills import --dry-run && sbx skills import --force
Start a sandbox with a published port and without shared skills, e.g. for an isolated web-service test.
$ sbx run -p 8080:8080 --no-share-skills my-image
  • Adds SSH access to sandboxes (experimental): run sbx setup ssh once to register a managed *.sbx SSH config entry, then connect to any sandbox with ssh <name>.sbx from OpenSSH clients or remote-dev tools like VS Code and Cursor.
  • Adds sbx skills import to discover and import host agent skills into a persistent store shared across sandboxes, with --dry-run to preview and --force to overwrite existing skills.
  • New --no-share-skills flag on sbx run / sbx create lets individual sandboxes opt out of the shared skills store.
  • Adds -p/--publish flag to sbx create and sbx run for publishing sandbox ports at creation time.
  • New DOCKER_SANDBOXES_PROXY=system env var routes sandbox egress through the host OS proxy (macOS/Windows), including PACPACA Proxy Auto-Configuration file, a JavaScript snippet a browser or OS fetches to decide which proxy server to route each request through, letting cyber tools intercept or redirect traffic without per-app configuration. auto-config URLs.
+1 more show less
  • Audit now emits execution-outcome records for network egress (per allowed connection) and filesystem mounts (per allowed path), including success, latency, and error class.

6 details in Everything

Was this useful?

Tencent Cloud CubeSandbox

Sources Release notes → v0.6.0 Notes

Instant, Concurrent, Secure & Lightweight Sandbox for AI Agents.

CubeSandbox v0.6.0 adds Kubernetes deployment, E2B-compatible volumes, template aliases, configurable Host headers, and node isolation.

└──▷ GET THIS VERSION
$ git clone --branch v0.6.0 https://github.com/TencentCloud/CubeSandbox.git
# already have the repo? check out this version:
$ git checkout v0.6.0
└──▷ TRY IT
Isolate a compute node before maintenance so no new sandboxes are scheduled to it while existing ones keep running.
$ cubemastercli node isolate <node-id>
Lift isolation on a node after maintenance is complete to allow new sandbox scheduling to resume.
$ cubemastercli node unisolate <node-id>
  • Deploy CubeSandbox control-plane and compute nodes on TKE, standard Kubernetes, or k3s via Helm Chart, with support for rolling upgrades and canary releases.
  • New E2B-compatible Volume framework lets operators plug in custom backend storage via Binary or 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. plugins, with Create/Destroy/Attach/Detach lifecycle hooks and sandbox–Volume binding APIs.
  • Template aliases can be set at creation time, allowing sandboxes to be launched by alias name instead of template ID.
  • New network.maskRequestHost parameter lets operators configure the Host header forwarded into a sandbox (e.g. localhost:${PORT}), improving compatibility with Host-based routing inside sandboxes.
  • New compute node isolation commands (cubemastercli node isolate / unisolate) let operators drain a node for maintenance without disturbing running sandboxes.
+9 more show less
  • CubeOps is now a standalone service decoupled from CubeAPI, making the control-plane API layer lighter and more extensible.
  • CubeMaster DAO layer now supports PostgreSQL as a metadata backend.
  • New CUBE_AUTO_MIGRATION environment variable allows skipping automatic DB schema migration at startup for controlled upgrade workflows.
  • Support deleting sandboxes that are in the paused state.
  • In-sandbox cgroup switches from v1 to v2.
  • CubeEgress MITM now writes subjectAltName on leaf certificates for broader TLS verification compatibility.
  • New cubemastercli volume-related subcommands added alongside Volume framework support.
  • Templates can now enable the ivshmem device.
  • Cubelet state tmpfs supports dynamic expansion.

14 details in Everything

Was this useful?
Detection & Response
◆  Endpoint Security (EDR/XDR)

Corellium

Sources Blog / feed → Release 7.12.0 (2026-07-23) Blog

Corellium is a virtualization platform enabling security researchers and developers to run iOS, Android, and Linux systems virtually for testing and analysis.

Corellium 7.12 adds Metal graphics support, KASAN/HWASAN kernel debugging, PongoOS pre-boot, MCP ServerMCP ServerA process that exposes tools, resources, and prompts over Anthropic's Model Context Protocol, letting an AI agent call external APIs or local services through a standardised interface. integration, and folder downloads.

└──▷ HOW TO FIND IT
Download an entire app data directory from a virtual device at once for forensic analysis or artifact collection, instead of extracting files one by one.
📍In the Corellium console, open the Files Browser for your virtual device, navigate to the target folder (e.g., /var/mobile/Containers/Data/Application/<app-uuid>), select the folder, and click Download to retrieve the full directory tree as a compressed archive.
  • Adds initial Apple Metal graphics support via a software-based implementation targeting Flutter applications, enabling apps that previously failed to launch or render in virtualized iOS environments.
  • Adds Android KASAN/HWASAN kernel debugging support, giving security researchers visibility into memory corruption, use-after-free, and buffer overflow issues at the kernel level.
  • Enables pre-boot of supported Apple devices directly into PongoOS before iOS loads, streamlining jailbreak research and low-level boot customization workflows.
  • Introduces an integrated Model Context ProtocolModel Context ProtocolModel Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. (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 as the first phase of next-generation API/SDK modernization, allowing AI tools and autonomous agents to interact directly with Corellium environments.
  • The Files Browser now supports downloading entire folders and nested sub-folder structures from virtual devices in a single operation.
+2 more show less
  • Adds support for iOS/iPadOS 26.5.2 and iOS/iPadOS 26.6.0 betas 2–4 and 27.0.0 betas 1–3 (beta OS access for Falcon Premium customers).
  • Updates Frida framework integration to support the latest Frida releases.

7 details in Everything

Was this useful?
◆  Detection Engineering

Elastic Detection Rules

Sources Release notes → dev-v2.0.0 Notes

Detection Rules are automated security patterns that identify suspicious activities and threats by analyzing event data in real-time.

Elastic Detection Rules dev-v2.0.0 adds MITRE ATT&CKMITRE ATT&CKA publicly maintained knowledge base of adversary tactics and techniques observed in real attacks, maintained by MITRE. Cyber tools use it as a common taxonomy to classify, map, and communicate threat behavior. v19 support and 80+ new rules across AWS, Azure, Linux, GCP, and network protocols.

└──▷ GET THIS VERSION
$ git clone --branch dev-v2.0.0 https://github.com/elastic/detection-rules.git
# already have the repo? check out this version:
$ git checkout dev-v2.0.0
  • Adds 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. ATT&CKATT&CKA publicly maintained knowledge base by MITRE cataloguing adversary tactics, techniques, and procedures observed in real attacks, giving cyber tools a common taxonomy for classifying and correlating threat behavior. v19 threat mapping support, automatically applied on Elastic Stack >= 9.5 with dynamic tag assignment.
  • New AWS rules: CloudTrailCloudTrailAWS service that records API calls and account activity across an AWS environment, giving cyber tools an audit log to detect anomalous or malicious actions. management event disabling, GuardDuty suppression detection, IAMIAMIdentity and Access Management — a framework of policies and services for controlling who can authenticate to systems and what resources they are authorized to use, letting cyber tools enforce least-privilege and audit access. console login without 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., IAM user login from multiple geolocations, account closure, org-leave attempts, Cognito unauthenticated identity pool credential issuance, Bedrock AgentCore execution role misuse, Bedrock prompt/completion credential detection, ECRECRAmazon Elastic Container Registry, AWS's managed service for storing and distributing Docker and OCI container images, giving cyber tools a private, IAM-gated registry without self-hosting infrastructure. public access grants, SageMaker suspicious lifecycle config, Lambda cross-account invocation, Lambda high-frequency invocation, Lambda event source mapping creation, Lambda layer external sharing, Lambda function URL public access, Lambda deletion, Lambda policy cross-account update, Backup vault deletion/lock removal, Backup recovery point deletion, 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. imported key material deletion, IAM inline policy added to group, IAM permissions boundary modification, IAM login profile creation/modification, and IAM account password policy deletion.
  • New Azure/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. rules: AKSAKSAzure Kubernetes Service, Microsoft's managed Kubernetes offering, which lets cyber tools offload cluster provisioning and scaling to Azure so engineers focus on workload logic instead of control-plane operations. API server proxying requests to Kubelet, ROPC authentication with unknown client ID, VM extension installation and suspicious child process via CustomScript, VM configuration modification, VM Managed Run Command with unusual principal, VM Extension CRUDCRUDThe four basic database operations — Create, Read, Update, Delete — used as shorthand for any system that manages persistent data records, giving cyber tools a standard model for interacting with stored information. from unusual ASNASNAutonomous System Number — a globally unique identifier assigned by regional internet registries to a network operator, letting routing protocols distinguish which organization controls a block of IP addresses., VM Boot Diagnostics retrieval, Serial Console connect with unusual user/ASN, Azure AD Graph access with unusual user/ASN, unusual client/user, suspicious user-agent, and 4xx error surge, Azure AD Graph potential enumeration via ROADrecon, Entra ID AiTM phishing-kit chain, device registration with phishing-kit default OS build, conditional access MFA bypass via Microsoft GraphMicrosoft GraphMicrosoft's unified REST API for accessing data and services across Microsoft 365, Azure AD, and related cloud products, giving cyber tools a single authenticated endpoint to query users, emails, files, and sign-in logs., multiple device registrations by single user, impossible travel sign-in, ROPC auth/legacy client detection, OAuth device code sign-in to Azure AD Graph enumeration, OAuth device code phishing, TAP creation, guest-to-member promotion, and OAuth redirect URI abuse.
  • New GCP/GKEGKEGoogle Kubernetes Engine, Google's managed Kubernetes service, which lets cyber tools run containerized workloads without managing the underlying cluster infrastructure. rules: GKE Kubernetes rule conversions (parts 1 and 2) adding coverage for GKE API and workload activity.
  • New Linux rules: kernel core_pattern container escape, systemd service override config file creation, Redis CONFIG SET SSH authorized key injection, potential Redis Lua activity, AMSI bypass via 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. NdrClientCall hook (PowerShell 4104), Linux LPE rule migrations from Endpoint Rules to Detection Rules (parts 1 and 2).
+6 more show less
  • New network protocol detection rules: ICMP tunneling to the internet, ICMP timestamp/information request from the internet, ICMP redirect from internal host, deprecated TLS version or weak cipher negotiated externally, DHCP starvation via high client MAC cardinality, multiple DHCP servers responding to the same transaction, potential SSH reverse port forwarding, DNS tunneling via nslookup, 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. activity from the internet, and protected storage service access via SMB.
  • New web/application rules: PHP file creation in WordPress plugin directory, potential SQL injectionSQL injectionAn attack that makes an application treat untrusted input as part of a database command. It can expose or modify data by changing what the query does. against Microsoft SQL Server, web server cloud metadata SSRFSSRFServer-side request forgery, an attack that tricks a server into making requests chosen by an attacker. Those requests can reach internal services the attacker cannot access directly. request, and Java dropped-and-executed with DNS lookup.
  • New Windows/endpoint rules: GenAI CLI started with unsafe permission bypass, Quick Assist full control sharing mode enabled, MS Teams rogue help desk detection, and Splunk Enterprise PostgreSQL Sidecar pre-auth RCERCERemote code execution, the ability to run attacker-chosen code on another system through a vulnerability. Its impact depends on the permissions and access that code receives. detection.
  • New Google Workspace rules: impossible travel login detection.
  • New M365 rules: unusual device code granting detection and Microsoft Defender 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. promotion rules.
  • Related integrations version constraints updated from ^ (caret/compatible) to >= (greater-or-equal) in emitted rule output.

11 details in Everything

└──▷ BREAKING ON UPGRADE
  • !Emitted related integrations now use >= instead of ^ for version constraints, which is expected to not function on Elastic stacks older than 9.5 — ensure your stack schema map and/or packages.yml match your stack version before upgrading.
  • !MITRE ATT&CK v19 mappings are automatically applied on stack >= 9.5; to use v19 on stacks <= 9.4 you must explicitly set DR_THREAT_MAPPING_FRAMEWORK="MITRE ATT&CK" and DR_THREAT_MAPPING_VERSION=1, or add threat_mapping_framework: "MITRE ATT&CK" and threat_mapping_version: "19" to your _config.yaml.
Was this useful?
Exposure & Vulnerability Management
◆  Attack Surface & Exposure Management

runZero

Sources Blog / feed → runZero 5.0.260723.0 (2026-07-23) Blog

runZero is a continuous asset intelligence platform that discovers, inventories, and monitors all connected devices and software across networks in real-time.

runZero 5.0.260723.0 adds Purdue level visibility for Dragos-enriched 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. assets.

  • Displays Purdue Model levels for assets sourced or enriched from Dragos integration, enabling OT/ICS network segmentation visibility.

1 detail in Everything

Was this useful?
Identity, Access & Data Protection
◆  Secure Access (VPN / SASE / ZTNA)

DNSFilter

Sources Changelog → changelog · Jul 24, 2026 Changelog

DNSFilter is a cloud-based DNS security platform that blocks malicious websites, filters unwanted content, and protects networks from cyber threats in real time.

DNSFilter redesigns the Query Log with multi-select filters, cross-org switching, and URL-persistent views, plus live local domain sync in the macOS client.

└──▷ HOW TO FIND IT
Investigate DNS activity across multiple organizations simultaneously without losing your filter context — select multiple Sites and Roaming Clients in the filter bar, then switch orgs to compare.
📍In the console, go to Query Log › top-level filter bar, select multiple Sites/Roaming Clients/Relays/Users, then use the org toggle to switch organizations while filters remain active.
Share a specific DNS investigation view with a teammate by bookmarking the URL after applying filters — active filters persist in the URL automatically.
📍In the console, go to Query Log, apply the desired filters and timeline via the custom timeline picker, then copy and share the resulting URL.
Set the Query Log as your default landing page so it opens immediately on login.
📍In the console, navigate to the Query Log page, then use the Set as Default Landing Page option to bookmark it as your homepage.
  • Redesigned DNS Query Log with a top-level filter bar supporting multi-select across Sites, Roaming Clients, Relays, and Users.
  • MSPMSPA managed service provider, a company that operates IT services for customers. It may manage systems, networks, backups or other ongoing technology needs. admins can now toggle between organizations directly from the Query Log without losing active filters, enabling faster cross-org investigation.
  • Introduced a custom timeline picker in the Query Log to narrow search windows down to the second.
  • Active Query Log filters now persist in the URL, enabling bookmarking and sharing of specific filtered views.
  • New Query Log data columns added: Resolved IPs, Query Type, and Roaming Client OS Type.
+3 more show less
  • Any page in the app can now be set as the default landing page upon login.
  • Live local domain and resolver sync: configuration changes in the dashboard now apply to the macOS Roaming Client without a daemon restart.
  • Automatic Registration toggle for Mobile Agent now displays its 7-day duration and explicit end date at a glance.

8 details in Everything

└──▷ BREAKING ON UPGRADE
  • !Starting with macOS Roaming Client v2.4.4, macOS 13 Ventura or later is required; macOS 12 Monterey is no longer supported.
Was this useful?

Zscaler Private Access

Sources Blog / feed → changelog · Jul 24, 2026 Blog

Zscaler Private Access provides zero-trust network access to internal applications without exposing them to the public internet.

ZPA gains Autonomous App Shield and Autonomous User-to-App Segmentation to continuously protect and isolate private apps.

  • New Autonomous App Shield continuously assesses private applications via App Connectors, reasons about per-app risk using global threat intelligence, and applies targeted protections automatically — adapting when developers ship new releases without manual re-tuning.
  • New Autonomous User-to-App Segmentation narrows user and third-party access to exactly the applications required, reducing 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. opportunity when a foothold is gained.
  • New Zscaler Lateral Threat Bundle packages application hiding (ZPA), autonomous segmentation, and App Shield into a unified offer for layered private-app defense.

3 details in Everything

Was this useful?

Slack Nebula

Sources Release notes →Source code → v1.11.0 Notes Code

A scalable overlay networking tool with a focus on performance, simplicity and security

Nebula v1.11.0 adds macOS network-change rebinding, IPv6IPv6The current version of the Internet Protocol, standardized by the IETF, using 128-bit addresses to replace the exhausted IPv4 space. Cyber tools target it because modern networks run dual-stack or IPv6-only. rejects, stdin/stdout cert I/O, and six breaking config changes.

└──▷ GET THIS VERSION
$ git clone --branch v1.11.0 https://github.com/slackhq/nebula.git
# already have the repo? check out this version:
$ git checkout v1.11.0
└──▷ USE IT
Keep macOS laptops connected when they roam between wifi and wired — disable the default rebind behaviour only if your environment requires it.
yaml
# In config.yml — omit or set to true to enable (the default)
listen:
  rebind_on_network_change: true
  • Adds automatic listener rebind and lighthouse re-query on macOS when the underlay network changes (wifi↔wired), controlled by listen.rebind_on_network_change (default true).
  • Generates IPv6 reject packets for blocked traffic, matching existing IPv4 reject behavior.
  • Accepts - as a path in nebula-cert subcommands to read from stdin or write to stdout.
  • Searches for both config.yml and config.yaml automatically in service and CLI modes.
  • Adds version labels to Docker/OCIOCIOpen Container Initiative standard, maintained by the Linux Foundation, defining the image format and runtime spec that container tools share so images built once run anywhere OCI-compliant. images.
+8 more show less
  • Signs Windows release binaries.
  • Supports port-free ICMP firewall rules — port is no longer required in ICMP rule definitions.
  • Tracks ICMP traffic per-flow via connection tracking.
  • Returns NODATA instead of NXDOMAIN from the embedded DNS server when a name exists but lacks the requested record type, fixing AAAA-first resolvers (busybox/Alpine).
  • Records the local Nebula host's own details in the DNS server.
  • Reloads the firewall automatically when unsafe networks in the certificate change, without requiring a restart.
  • Triggers an immediate lighthouse update when reconnecting to or adding a lighthouse instead of waiting for the next tick.
  • Adds Control.Wait, a new API method that blocks until Nebula has fully stopped and returns the first fatal reader error.

13 details in Everything

└──▷ BREAKING ON UPGRADE
  • !Logging has switched from logrus to Go's structured slog: log levels are now upper case (level=INFO), trace prints as level=DEBUG-4, timestamps are always RFC3339Nano, logging.timestamp_format is ignored, and some messages were reworded. Embedders must update constructors to accept *slog.Logger.
  • !firewall.inbound_action and firewall.outbound_action were each applied to the opposite direction; this is now corrected. If you set either, you are currently getting the behavior of the other and likely need to swap them before upgrading.
  • !On Windows, Nebula now installs WFP PERMIT filters for the nebula adapter and listener port by default, bypassing Windows Defender Firewall (WDF) inbound rules for those. Set tun.windows_bypass_wdf and listen.windows_bypass_wdf to false to leave WDF in charge.
  • !On Windows, the Nebula device network category is now set to private instead of whatever Windows chose (usually Public). Set tun.network_category to unset to keep the old behavior.
  • !Reject packets for non-TCPTCPA network protocol that delivers a reliable, ordered stream of bytes between two endpoints. It detects lost data and retransmits it. traffic now use ICMP code 13 (communication administratively prohibited) instead of code 3 (port unreachable). Anything keying off the old code needs updating.
  • !The SSH debug server's profiling commands (start-cpu-profile, save-heap-profile, save-mutex-profile) are now confined to sshd.sandbox_dir (default $TMP/nebula-debug). Absolute paths outside that directory are rejected; the directory is not created automatically. Scripts using paths elsewhere must set sshd.sandbox_dir accordingly.
Was this useful?
◆  Identity & Access Management

iDenfy

Sources Changelog → changelog · Jul 24, 2026 Changelog

iDenfy is an identity verification platform that uses AI-powered document scanning and biometric facial recognition to verify user identities and prevent fraud.

  • Supports shared identity verification across companies in 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. (Know Your Business) workflows.
Was this useful?

Clerk

Sources Changelog → changelog · Jul 24, 2026 Changelog

Clerk is a user authentication and identity management platform that provides developers with pre-built UI components, APIs, and SDKs for sign-up, sign-in, and user management.

Clerk CLI gains a clerk mcp command group for one-command 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 registration across 10 AI clients, plus configurable default OAuth scopes.

└──▷ TRY IT
Register the Clerk MCP server across all detected AI clients in one shot — useful for onboarding a new machine or CI environment.
$ clerk mcp install --all
Target only Claude Code and Cursor for MCP registration when you don't want to touch other clients on the machine.
$ clerk mcp install --client claude --client cursor
Verify your Clerk MCP server is reachable (not just installed) after registration, to catch config or network issues early.
$ clerk doctor
  • Adds clerk mcp install command that detects supported AI clients (Claude Code, Cursor, GitHub Copilot, Windsurf, Gemini CLI, Codex, opencode, OpenClaw, Warp, Hermes Agent) and registers the Clerk MCP server in each, user-globally, in a single interactive step.
  • Adds --client and --all flags to clerk mcp install for non-interactive, scripted registration targeting specific or all supported clients.
  • Adds clerk mcp run, a built-in stdio↔Streamable-HTTP bridge that replaces the npx mcp-remote dependency; updating the CLI updates the bridge everywhere with no re-install.
  • Adds clerk mcp list to display every Clerk MCP entry registered across all clients, and clerk mcp uninstall to remove them selectively.
  • Extends clerk doctor to probe the configured Clerk MCP server with a real initialize handshake, distinguishing not installed, installed but unreachable, and working states.
+1 more show less
  • Supports configuring default OAuth scopes for dynamic client registration via the OAuth Applications settings in the Clerk Dashboard or the Backend API, preventing invalid_scope errors when clients omit the scope parameter.

6 details in Everything

└──▷ BREAKING ON UPGRADE
  • !Beginning January 18, 2027, CBC-mode 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./TLS cipher suites (including TLS_ECDHE_ECDSA_WITH_AES_128_CBC_SHA, TLS_ECDHE_RSA_WITH_AES_256_CBC_SHA, TLS_RSA_WITH_AES_256_CBC_SHA, and others) will be removed from Clerk-managed subdomains (clerk.example.com, accounts.example.com); clients that can only negotiate a CBC cipher will fail the TLS handshake after their subdomain's certificate rotates.
Was this useful?

Ping Identity

Sources Changelog → changelog · Jul 24, 2026 Changelog

Ping Identity is an identity security platform that provides authentication, single sign-on, multi-factor authentication, and access management solutions for enterprises.

Ping Identity AIC adds email scripting, secrets in nodes, Rich Authorization Requests, 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. metadata import options, and a new validate-secrets endpoint.

  • Adds emailService binding for next-generation scripts, enabling email delivery directly from journey scripts.
  • New realm-level OAuth2OAuth2An open authorization framework (RFC 6749) that lets a service grant limited access to a user's resources on another service without exposing credentials, giving cyber tools a standard way to authenticate against third-party APIs. Provider setting enables Rich Authorization Requests (RFC 9396) processing when an RCSRCSRich Communication Services, a messaging protocol standard maintained by the GSMA that replaces SMS with richer features like read receipts, file transfer, and group chat over carrier or IP networks. is configured, and controls advertisement of authorization_details_types_supported on the well-known endpoint.
  • Individual 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. clients can now be configured with Authorization Detail Type(s) to restrict which authorization detail types each client may request.
  • Dynamic client registration requests now support authorization_details_types to declare the authorization detail types a client supports at registration time.
  • SAML metadata import now supports skipping signature verification.
+12 more show less
  • Introduces Patch Object node v2.0, which always updates managed identity objects as Advanced Identity Cloud (v1.0 deprecated but still functional).
  • OAuth 2.0 request state is now stored in CTS (Core Token Service) instead of an oversized browser cookie, preventing large requests (e.g., with big id_token_hint) from failing after authentication.
  • Custom report parameters can now be marked as optional, providing more flexibility when running reports.
  • Proxy Connect enabled tenants can now manage Proxy Connect configuration directly in the Advanced Identity Cloud admin console without using the API.
  • Hosted journey pages now display hCaptcha text and challenges in the end user's configured locale.
  • Column visibility preferences for data tables in the admin console and hosted account pages are now customizable and saved between sessions.
  • Super administrators can now configure the redirect URI to use a custom domain when setting up federated administrator access.
  • Social 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. can now be created and edited directly in the Advanced Identity Cloud admin console.
  • Self-service journey mappings can now be configured directly in the Advanced Identity Cloud admin console.
  • The promotions API now returns report metadata ordered by creation date, most recent first.
  • Journey annotation notes now display multi-line formatting in view mode.
  • Page node v3 is now available, generating a new CAPTCHA token for each validation request when both CAPTCHA and live password validation are used together.

17 details in Everything

Was this useful?
◆  Data Security & DLP

Immuta

Sources Release page → updated · Jul 24, 2026 Notes

Immuta is a data security platform that automates access control, privacy policy enforcement, and compliance monitoring across cloud data environments.

└──▷ HOW TO FIND IT
When onboarding a data product with sensitive columns, configure distinct request forms so access requestors and masking-exception requestors each see only relevant questions and agreements.
📍In the Immuta UI, navigate to your data product › Settings › Request Forms, then configure a separate form under Data Access Requests and another under Masking Exception Requests, adding tailored questions and data use agreements to each.
Write a subscription policy that restricts AI-restricted data to human users only, excluding all agent accounts system-wide.
📍In the Immuta UI, create a subscription policy targeting your data source, set the condition to allow only members of the built-in group All Users (excluding All Agents), and save.
Enable 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 the Alation integration to avoid recurring token expiry issues.
📍In the Immuta UI, navigate to Settings › Integrations › Alation, select OAuth 2.0 as the authentication method, and provide the required OAuth credentials.
  • Adds separate request forms for data products: one for data access requests and one for masking exception requests, each with independent questions and data use agreements.
  • Expands Snowflake manual policy resync to cover both data policies and subscription policies simultaneously.
  • Adds automated sensitive data discovery (regex, dictionary, and column-name identifiers) for the Amazon Redshift viewless integration.
  • Adds 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 support to the Alation integration, eliminating repeated token refresh cycles required by API key authentication.
  • Embeds Immuta documentation and an AI-powered chatbot directly in the Immuta UI via a new Documentation icon in the top menu.
+1 more show less
  • Starburst (Trino) principal name matching is now case-insensitive, enabled by default for all customers as of July 1.
└──▷ BREAKING ON UPGRADE
  • !Snowflake integration architecture update (rolling out July 23, 2026): Immuta policy role names used to grant users access to Snowflake data sources will be changed and recreated; any automation or tooling that references specific policy role names by their current values will need to be updated.
  • !Immuta will no longer display integration error banners when integration validation fails for Snowflake; errors will only be visible through the connections experience.
  • !When a data source is disabled in Immuta, Immuta now revokes data owners' Immuta-managed grants in the data platform (behavior change releasing 8/1 by default, 9/1 for all).
Was this useful?
Governance, Risk & Compliance
◆  Integrated & IT Risk Management

CISO Assistant

Sources Release notes → powerbi-v1.0.1 Notes

CISO Assistant is a one-stop-shop GRC platform for Risk Management, AppSec, Compliance & Audit, TPRM, BIA, Privacy, and Reporting. It supports 150+ global frameworks with automatic control mapping, including ISO 27001, NIST CSF, SOC 2, CIS, PCI DSS, NIS2, DORA, GDPR, HIPAA, CMMC, and more.

CISO Assistant gains a native Power BI connector for direct BI integration.

└──▷ GET THIS VERSION
$ git clone --branch powerbi-v1.0.1 https://github.com/intuitem/ciso-assistant-community.git
# already have the repo? check out this version:
$ git checkout powerbi-v1.0.1
Was this useful?
Build
◆  AI Coding Agents

Kiro

Sources Release page → V2 to V3 Agent Migration Notes

Kiro is an AI-powered IDE by AWS that uses "specs" to translate requirements into structured plans, then automatically generates and implements code.

Kiro V3 adds /upgrade-agent command to migrate V2 custom agent configs to the universal format

└──▷ TRY IT
Convert existing V2 custom agent configs in place so they work in both V2 and V3 sessions.
$ /upgrade-agent
  • New /upgrade-agent command migrates V2 custom agent configuration files to the universal format compatible with both V2 and V3 modes.
  • Automatic retries for interrupted model responses, reducing manual intervention during long agent sessions.

2 details in Everything

Was this useful?

Pi

Sources Release notes → v0.82.0 Notes

AI agent toolkit: unified LLM API, agent loop, TUI, coding agent CLI

Pi v0.82.0 adds constrained tool sampling, OAuth login for 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./Kimi Code, and streaming bash 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. events.

└──▷ GET THIS VERSION
$ git clone --branch v0.82.0 https://github.com/earendil-works/pi.git
# already have the repo? check out this version:
$ git checkout v0.82.0
└──▷ TRY IT
Authenticate with OpenRouter interactively so Pi can use it without manually setting an API key.
$ /login
Inspect session and model context from inside a bash tool script — useful for writing tools that behave differently per provider or model.
$ echo "Session: $PI_SESSION_ID | Provider: $PI_PROVIDER | Model: $PI_MODEL | Reasoning: $PI_REASONING_LEVEL"
  • Adds Tool.constrainedSampling supporting strict JSON Schema (prefer/require) and OpenAI Lark/regex grammar variants across OpenAI, Anthropic, Amazon Bedrock, Google Gemini, and Mistral providers.
  • Adds model capability flags supportsGrammarTools and supportsStrictTools to gate constrained sampling requests to models that support them.
  • Adds OAuth PKCEPKCEProof Key for Code Exchange, an OAuth safeguard that binds an authorization code to the client that requested it. An intercepted code cannot be redeemed without the client's matching verifier. login for OpenRouter via /login, automatically minting a user-controlled API key without manual configuration.
  • Adds OAuth device-authorization login for Kimi Code subscriptions via /login, with automatic token refresh.
  • Exposes PI_SESSION_ID, PI_SESSION_FILE, PI_PROVIDER, PI_MODEL, and PI_REASONING_LEVEL as environment variables inside commands run by built-in and factory-created bash tools.
+1 more show less
  • Adds streaming bash_execution_update RPC events for direct bash commands, correlated with request IDs.

6 details in Everything

Was this useful?

OpenAI Codex CLI

Sources Release notes →Source code → rust-v0.146.0-alpha.6 3 releases · 2026-07-23 → 2026-07-24 Notes Code Pre-release

Lightweight coding agent that runs in your terminal

Codex CLI alpha.6 adds remote code-mode host support, nonblocking 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. interrupts, and skill-catalog context warnings.

└──▷ GET THIS VERSION
$ git clone --branch rust-v0.146.0-alpha.6 https://github.com/openai/codex.git
# already have the repo? check out this version:
$ git checkout rust-v0.146.0-alpha.6
└──▷ TRY IT
Connect the app-server to a remote code-mode host over a secure 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. instead of the default in-process host.
$ codex-app-server --code-mode-host wss://example.com/code-mode --listen ws://127.0.0.1:9000
  • Adds --code-mode-host flag to the app-server to connect to a remote code-mode host over WebSocket (wss://).
  • Supports WebSocket transport for the code-mode host in the app-server, enabling remote code-mode sessions.
  • Exposes remote skill icon URLs through the app server.
  • Exposes Browser Use requirements through the app server.
  • Routes exec-server HTTP and WebSocket traffic through the configured proxy policy.
+7 more show less
  • Adds trusted plugin script attribution, tracking which plugin scripts originated command executions.
  • Registers the Guardian V2 feature flag.
  • Allows disabling the update_plan tool via configuration.
  • Preserves refreshed Apps tools across 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. runtime updates.
  • Keeps side conversations open when switching threads.
  • Supports incremental replay of updated thread items.
  • Enforces writer ownership for thread archive and deletion operations.

12 details in Everything

0 more releases in this issue 2 more releases in this issue · 2026-07-23 → 2026-07-24
rust-v0.146.0-alpha.3.1 Notes Code Pre-release

Custom model providers can now opt into standalone web search via a new supports_standalone_web_search setting.

└──▷ GET THIS VERSION
$ git clone --branch rust-v0.146.0-alpha.3.1 https://github.com/openai/codex.git
# already have the repo? check out this version:
$ git checkout rust-v0.146.0-alpha.3.1
  • Adds supports_standalone_web_search model-provider setting that lets custom Responses API providers opt into the standalone web.run web search tool.
  • Routes standalone web search requests through the custom provider's own endpoint and authentication when the setting is enabled.

2 details in Everything

rust-v0.146.0-alpha.5 Notes Code Pre-release

Codex CLI alpha.5 adds thread pinning, 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 prefix omission, custom provider web search, and MCP connection reuse.

└──▷ GET THIS VERSION
$ git clone --branch rust-v0.146.0-alpha.5 https://github.com/openai/codex.git
# already have the repo? check out this version:
$ git checkout rust-v0.146.0-alpha.5
  • Adds isPinned field to thread responses with pin/unpin support via thread/metadata/update and an isPinned filter on thread/list, enabling persistent thread pinning with cursor-based pagination.
  • Allows omitting MCP tool name prefixes on a per-server basis, reducing verbosity when working with trusted MCP servers.
  • Allows custom providers to opt into standalone web search, extending web-search capability beyond first-party models.
  • Reuses MCP connections across runtime refreshes and replaces closed connections during reconciliation, reducing reconnection overhead.
  • Allows disabling the multi-agent wait tool, giving operators control over multi-agent coordination primitives.
+4 more show less
  • Infers the bundled Claude Code plugin marketplace, enabling automatic plugin discovery for Claude Code users.
  • Uses the API plugin marketplace for Amazon Bedrock, broadening plugin availability for Bedrock-backed deployments.
  • Sets a default User-Agent header for MCP HTTP requests, improving server-side auditability of MCP traffic.
  • Caches remote plugin catalogs by scope, reducing repeated fetches when working with scoped plugin registries.

9 details in Everything

Was this useful?

Cline

Sources Release notes →Source code → desktop-v0.0.4 Notes Code

Autonomous coding agent as an SDK, IDE extension, or CLI assistant.

Cline desktop v0.0.4 adds workspace-free chat, file drag-and-drop, one-time routines, and agentic compaction by default.

└──▷ GET THIS VERSION
$ git clone --branch desktop-v0.0.4 https://github.com/cline/cline.git
# already have the repo? check out this version:
$ git checkout desktop-v0.0.4
└──▷ TRY IT
Create a headless scheduled routine that runs unattended in YOLO mode — no human approval prompts.
$ cline schedule create "Nightly dependency audit" --cron "0 2 * * *" --prompt "Check for outdated dependencies and open a PR if any are found" --workspace /path/to/repo --mode yolo
  • Supports chat sessions without opening a project folder (workspace-free mode).
  • Drag and drop files directly onto the chat to attach them.
  • Image attachments display inline in the chat transcript.
  • Schedule one-time (non-recurring) routines with navigation to jump to a routine's run.
  • Agent sessions now use agentic compaction by default to keep long conversations within context.
+6 more show less
  • Headless routines now default to YOLO mode (--mode yolo) for unattended execution.
  • New --mode <act|plan|yolo> option added to schedule create and schedule update commands.
  • New first-run onboarding flow on launch.
  • New custom overlay title bar with in-app navigation.
  • Redesigned channel setup as expandable cards.
  • Added a setting to replay the new-user experience.

11 details in Everything

Was this useful?

Goose

Sources Release notes →Source code → v1.44.0 Notes Code

an open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM

Goose v1.44.0 adds new model support (GPT-5.6, Gemini, Muse Spark 1.1), new providers, --edit session flag, per-message usage stats, and OllamaCloud integration.

└──▷ GET THIS VERSION
$ git clone --branch v1.44.0 https://github.com/aaif-goose/goose.git
# already have the repo? check out this version:
$ git checkout v1.44.0
└──▷ TRY IT
Edit a prior session's conversation before forking it — useful for pruning context or replaying with adjusted prompts.
$ goose session --edit
  • Adds support for latest Gemini models.
  • Adds Muse Spark 1.1 support via the Meta Models API.
  • Adds MiniMax-M3 and missing M2.7 model variants.
  • Adds declarative Sakana AI provider for the OpenAI-compatible Fugu API.
  • Adds iFlytek Spark and Astron MaaS providers.
+10 more show less
  • Adds Fireworks AI as a declarative provider.
  • Adds OllamaCloudProvider with dynamic model discovery and context limits.
  • Adds 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. request parameters support.
  • New --edit session flag to edit a conversation before forking it.
  • New per-message usage stats UI showing tokens, cost, TTFTTTFTTime to first token, the wait between submitting a request and receiving the first generated token. It includes delays such as queuing and processing the prompt., and tokens/second.
  • Chunks command-classifier input with overlapping windows for improved security robustness.
  • Adds search filter for the provider grid in the UI.
  • Adds delete support for custom apps from the Apps UI.
  • Reconnects desktop ACP sessions automatically after sleep or connection loss.
  • Adds French, German, Italian, Portuguese, Indonesian, Malay, Vietnamese, and zh-TW desktop locales.

15 details in Everything

Was this useful?
Deploy
◆  Local LLM Runtimes

Ollama

Sources Release notes →Source code → v0.32.3 Notes Code

Get up and running with Kimi-K2.6, GLM-5.2, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.

Ollama v0.32.3 adds CUDACUDANVIDIA's platform and programming interface for running general computation on its GPUs. Nearly every AI training and inference stack is built on it first. on Windows ARM64ARM64The 64-bit instruction set architecture from ARM Holdings, used in modern mobile chips, Apple Silicon, and cloud server CPUs; tools target it to run natively on those platforms without emulation overhead., B200/CUDA 12 support, Laguna 2.1 model capabilities, and lower iGPU memory use.

└──▷ GET THIS VERSION
$ git clone --branch v0.32.3 https://github.com/ollama/ollama.git
# already have the repo? check out this version:
$ git checkout v0.32.3
  • Adds CUDA support on Windows ARM64 devices.
  • Adds NVIDIA B200 GPUGPUA graphics processing unit, a processor that performs many calculations in parallel. AI workloads use GPUs to speed up model training and inference. support via CUDA 12.
  • Reduces memory usage on Linux CUDA and ROCmROCmAMD's open software stack for running AI and compute workloads on its GPUs, the counterpart to NVIDIA's CUDA. integrated GPUsGPUsA graphics processing unit, a processor that performs many calculations in parallel. AI workloads use GPUs to speed up model training and inference..
  • Adds chat, thinking, and 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. support for Laguna 2.1 models.
  • Restores Claude Code Channels support when launching via ollama launch claude.

5 details in Everything

Was this useful?
browse all tools → score card for your toolchain →