This Week's Highlights — issue 018, October 30, 2026
THE CYBER TOOLCHAIN
NO. 018
The best tooling updates that shipped this week.
The week in numbers · Oct 23–29
Changes per week, last 12 weeks · Oct 23–29
Changes we read in that week's dailies, including docs and API changes. A gap means no daily was published.
Every feature release from the 1309 tools on our watchlist goes into the daily newsletter. Once a week we read the whole field side by side and grade each release from 1 to 5 on three things, weighted equally:
- Depth. How complete is the best new capability?
- Impact. How much does it change what you can do?
- Novelty. Has any other tool done this before? We check every capability against the history of every tool we track. Features most tools already have earn nothing here. Being first earns the most.
A tool is judged on everything it shipped that week, so a project that releases daily gets credit for the sum — and still only takes one slot. We would rather run a short list than a padded one.
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.
This issue came out before editions were ranked, so it shows the Standard edition in issue order.
The picks in full
External Network Pentesting sends AI agents to map your perimeter and exploit it from the outside — discovering subdomains, ports, and services, then chaining findings across hosts into a full attack path, with continuous retesting of newly discovered subdomains. No code access or credentials required.
Aikido Security launched External Network Pentesting for internet-exposed assets, added a Cloud Security Audit capability for cloud/IaCIaCInfrastructure as Code: the practice of defining and provisioning compute infrastructure through machine-readable config files rather than manual processes, giving cyber tools repeatable, version-controlled environment setup. attack-path analysis, achieved FedRAMP ModerateFedRAMP ModerateA US government authorization tier requiring cloud services to meet a defined set of NIST 800-53 controls before federal agencies can use them, giving cyber tools a pre-vetted compliance baseline for government contracts. authorization, and rounded out the release with new API endpoints, a 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. SIEMSIEMSecurity information and event management, collecting logs from many systems and correlating events to find suspicious activity. Analysts use it to investigate alerts and incidents. integration, expanded Zen Firewall language support, and Code Quality/Terraform improvements.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
what's the number?
Each feature carries 0–100 for how completely the vendor documented it — not how big or important the work is. A major capability described in eight words scores low, and that is the finding.
- depth0–40 what it does and how it works or what changed
- specificity0–30 names real surfaces — APIs, flags, formats, limits, numbers
- actionability0–30 enough to go use it — a named endpoint, flag, or config key tops this; a UI path is a starting point
Code and config rank first by construction: a documented endpoint, flag or config key scores at the top of specificity and actionability, so it sorts above a feature described only in prose. Under 60 the notes go thin — everything below the “thinner coverage” line is thinner documentation, not smaller work. Hover any meter for that feature's three sub-scores.
AI agents discover subdomains, open ports, and services from a list of top-level domains, then attempt exploitation using known CVEs, misconfigurations, and exposed secrets — including forgotten subdomains and open admin panels, with evidence-confirmed findings.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Agents chain findings across hosts, pivoting using credentials or access found on one system, and the report shows the full attack pathattack pathA possible sequence of steps an attacker could take to reach a valuable target. It connects weaknesses, permissions and reachable systems rather than treating each finding separately. from entry point to impact; the Agents tab shows each agent's current thinking, tools in use, and findings in real time.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Testing can run in continuous mode so newly discovered subdomains are tested shortly after they go live, with intensity controlled via a maximum requests-per-second per target and optional restriction to inside or outside business hours. A scoping agent checks how much of each subdomain you own, limits testing to owned parts, and routes all agent traffic through a proxy that blocks out-of-scope connections. The output is an auditor-ready report in the same format as Aikido's web app and mobile pentests, covering the 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. up to the login page.
Adds API - List CWE descriptions to retrieve CWECWECommon Weakness Enumeration, a catalog of recurring software and hardware weakness types. It describes classes of mistakes, such as injection, rather than individual vulnerability instances. metadata, API - List pending invites to query outstanding user invitations, API - List AI code analysis assessments to fetch AI code analysis results, API - List pentest assessments to retrieve pentest findings, API - Disable user to deactivate users, and API - Remove attach surface subdomain to remove subdomains from the attack surface.
New integration to optionally forward Aikido Activity logs to Google SecOps as a SIEM destination, configured via Settings > Integrations > SIEM by enabling Transmit Activity logs and selecting Google SecOps.
New feature where AI agents reason across cloud and IaC configurations to surface multi-step attack pathsattack pathsA possible sequence of steps an attacker could take to reach a valuable target. It connects weaknesses, permissions and reachable systems rather than treating each finding separately. and link each finding back to the causative template.
Adds Terraform Container management resources and publishes the Aikido provider to the OpenTofu registry.
Removes the maximum-rule restriction for Code Quality scans and enables auto-creation of PRs for issues found in full repository scans.
Adds Zen Firewall agent releases for Dotnet, Python, PHP, Ruby, and Node.
Aikido for Government achieves FedRAMPFedRAMPA US government program that standardizes cloud security authorization, letting vendors get a single assessment accepted across federal agencies instead of repeating the process for each one. Moderate authorization and is listed on the FedRAMP Marketplace, allowing federal agencies to adopt Aikido without a full security review.
Adds Hugging Face to supported malware ecosystems in Aikido Intel.
Cyberark is a protect your organization with Idira®. Modernize identity governance, automate access reviews, and secure every identity across your hybrid cloud environment.
Idira Agent Identity Security reaches GA, issuing SPIFFE SVIDs and short-lived OAuth 2.1 credentials to each AI agent and enforcing tool-level least-privilege on MCP servers, with zero standing privilege for databases and an onboarding flow for shadow agents.
Idira Agent Identity Security is now GA with Prisma AIRS, delivering per-agent SPIFFESPIFFEAn open standard, maintained by the CNCF, for assigning cryptographic identities to workloads in dynamic infrastructure so services can authenticate each other without passwords or static credentials./OAuth 2.1OAuth 2.1A consolidated draft of the OAuth authorization framework that merges OAuth 2.0 with its most important security BCP updates, dropping unsafe flows so tools can delegate user access without handling credentials directly. credentials, 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-level ZSP, on-behalf-of audit logging, shadow agent onboarding, and an activity map.
- New supported platformProvisions SPIFFE Verifiable Identity Documents (SVIDs) and 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..1 client credentials for each registered agent, issuing short-lived, time-bound access tokensaccess tokensA credential that authorizes a client to use a service or API. It usually has limited permissions and a lifetime; anyone holding it may be able to use those permissions. in place of long-lived shared secrets.
- Applies fine-grained access policies to specific tools available on an MCP server, enabling least-privilege and zero standing privilege (ZSP) controls per agent.
- New supported platformDelivers ZSP for databases via the Secure Infrastructure Access MCP server, with ZSP for cloud providers listed as coming soon.
- New integrationSupports automatic OAuth app registration for MCP servers that implement Dynamic Client Registration (DCR), capturing audit data for tool interactions.
- Brings shadow AI agents into central governance through an unmanaged agent onboarding flow, including agents identified by cloud-provider origin IDs such as AWS ARNs or App IDs.
+4 more show less
- Provides an interactive activity map for visual tracing, analysis, and audit of complex agent execution paths across the enterprise (early access).
- Logs every agent action with both the agent identity and the associated human identity (on-behalf-of), tracking which tool ran, on whose behalf, and through which MCP server.
- New integrationIntegrates Idira as the agentic identity provideridentity providerA service that authenticates users and supplies identity information to other applications. It is the sign-in authority behind many single sign-on systems. (IdPIdPA service that authenticates users and supplies identity information to other applications. It is the sign-in authority behind many single sign-on systems.) for the Prisma AIRS AI Agent Gateway, now generally available to AI Gateway customers.
- Includes secrets manager licensing with Idira Secure Agents licenses, consolidating secrets management for agent workloads.
9 details in Everything
Clearance enforces per-workload outbound and lateral containment by workload identity across clouds, Kubernetes, and serverless — down to individual containers, databases, and AI agents — with a shadow-rule monitor mode that proves exactly which connections a policy would block before any enforcement.
Aviatrix launched Clearance, a per-workload outbound and lateral containment system built on an autonomous policy-drafting agent and a monitor-mode testing workflow, alongside a free assessment and promotional pricing window.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
what's the number?
Each feature carries 0–100 for how completely the vendor documented it — not how big or important the work is. A major capability described in eight words scores low, and that is the finding.
- depth0–40 what it does and how it works or what changed
- specificity0–30 names real surfaces — APIs, flags, formats, limits, numbers
- actionability0–30 enough to go use it — a named endpoint, flag, or config key tops this; a UI path is a starting point
Code and config rank first by construction: a documented endpoint, flag or config key scores at the top of specificity and actionability, so it sorts above a feature described only in prose. Under 60 the notes go thin — everything below the “thinner coverage” line is thinner documentation, not smaller work. Hover any meter for that feature's three sub-scores.
Aviatrix Clearance™, part of the Aviatrix Containment Platform, adds Outbound Containment (limiting each workload to outbound destinations it is explicitly cleared for, named by domain rather than network address) and Lateral Containment (restricting workload-to-workload reachability by identity across clouds and Kubernetes/serverless paths not governed by hub-based inspection).︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ It enforces per-workload policy by identity and protocol across clouds, down to individual containers, serverless functions, databases, and AI agents.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
A free, read-only Containment Assessment at aviatrix.ai/containment-assessment installs nothing and can reserve AWS Marketplace enrollment. Clearance (Outbound Containment, Lateral Containment, and Intrusion Prevention, no workload cap) is free for all enterprises from November 1, 2026 through May 1, 2027 via AWS Marketplace. A General Availability tier offers the Containment Assessment plus 15 workloads in monitor or enforce mode with no time limit, priced per workload, on Aviatrix Cloud or Aviatrix Enterprise; Aviatrix Active Defense is available today in free early access for design partners.
A new Shadow Rule Set monitor mode evaluates a proposed policy alongside the live policy against real traffic and shows exactly which connections would be blocked, letting teams validate a containment policy before enforcing it.
Aviatrix Active Defense™ is an autonomous agent layer that models the estate from live traffic, drafts per-application policies without a tagging project or inventory import, proves each policy in monitor mode on real traffic, and keeps policies current as the estate changes.
Every containment policy can be turned on with a single setting and rolled back with a single action, covering enforcement and backout as one-step operations.
DTLS 1.3 per RFC 9147 lands in the most widely used crypto toolkit, alongside RFC 8701 GREASE and AVX-512/VAES post-quantum and AES-CBC performance work.
OpenSSL 4.1.0 adds DTLS 1.3DTLS 1.3A UDP-based variant of TLS 1.3, standardized by the IETF, that adds record-layer sequencing and retransmission to handle packet loss, giving tools encrypted low-latency channels without TCP overhead. (RFC 9147RFC 9147An IETF standard specifying DTLS 1.3, the datagram transport variant of TLS 1.3, giving tools encrypted, low-latency communication over UDP without the overhead of a TCP connection.) and GREASEGREASEA TLS extension mechanism (RFC 8701) maintained by the IETF that injects reserved dummy values into handshakes to keep implementations from rejecting unknown extensions, ensuring future protocol additions deploy cleanly. (RFC 8701RFC 8701An IETF specification that reserves specific TLS and QUIC extension values as "GREASE" placeholders, letting implementations send random reserved values to ensure peers tolerate unknown extensions without breaking.) support, new IKEv2 KDFIKEv2 KDFThe key derivation function specified in IKEv2 (RFC 7296), used to stretch shared secrets into keying material for IPsec tunnels, giving cyber tools a standard way to generate session keys during VPN negotiation. and DTLS listener APIs, and performance optimizations for ML-DSA/ML-KEM post-quantum crypto, alongside breaking changes to removed Configure options, the tsget utility's curl dependency, and dropped Windows build targets.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
what's the number?
Each feature carries 0–100 for how completely the vendor documented it — not how big or important the work is. A major capability described in eight words scores low, and that is the finding.
- depth0–40 what it does and how it works or what changed
- specificity0–30 names real surfaces — APIs, flags, formats, limits, numbers
- actionability0–30 enough to go use it — a named endpoint, flag, or config key tops this; a UI path is a starting point
Code and config rank first by construction: a documented endpoint, flag or config key scores at the top of specificity and actionability, so it sorts above a feature described only in prose. Under 60 the notes go thin — everything below the “thinner coverage” line is thinner documentation, not smaller work. Hover any meter for that feature's three sub-scores.
The no-ecdsa and no-ecdh Configure options are removed; use no-ec instead to disable elliptic curve support.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
The tsget utility now requires Net::Curl::Easy (from the Net-Curl CPANCPANThe Comprehensive Perl Archive Network, a public repository of reusable Perl modules maintained by the Perl community, giving cyber tools instant access to thousands of tested libraries without reinventing common functionality. distribution) instead of the abandoned WWW::Curl::Easy — install it before upgrading.
Adds optimized ML-DSAML-DSANIST's main post-quantum digital signature standard (FIPS 204), formerly Dilithium. It is the planned replacement for RSA and ECDSA signatures on certificates and code. and ML-KEMML-KEMNIST's post-quantum key-exchange standard (FIPS 203), formerly Kyber. Two parties use it to agree on a shared secret, usually alongside X25519 in a hybrid TLS handshake. NTT operations on ppc64le; optimized ML-DSA operations on x86_64 including AVX-512-optimized SHAKE x4 for ML-DSA and AVX-512/VAES optimizations for AES-CBC decryption.
OpenSSL 4.1.0 adds DTLSDTLSA variant of TLS adapted for UDP, standardized by the IETF, that adds handshake reliability and replay protection without sacrificing datagram semantics — letting cyber tools encrypt low-latency or connectionless traffic. 1.3 support per RFC 9147 (documented in the new ossl-guide-dtlsv13(7) manual page) and adds DTLS listener API support.
Adds VC-WIN32-MSVC2013 and VC-WIN64A-MSVC2013 build targets for MSVC 2013 C99 bridging support.
The VC-WIN64I (Windows-on-Itanium) and VC-CE (Windows CE) build targets are dropped from Configurations.
Adds RFC 8701 GREASE (Generate Random Extensions And Sustain Extensibility) support.
Adds initial support for the Elbrus2000 (e2k) architecture.
OpenSSL 4.1.0 adds IKEv2IKEv2The second version of the Internet Key Exchange protocol, standardized by the IETF, used to negotiate and authenticate IPsec tunnels. Cyber tools build on it for reliable, fast VPN session setup with built-in NAT traversal and EAP support. KDF (key derivation function) support.
- !The
no-ecdsaandno-ecdhConfigure options are removed; useno-ecto disable elliptic curve support. - !The
tsgetutility now requiresNet::Curl::Easy(from the Net-Curl CPAN distribution) instead of the abandonedWWW::Curl::Easy— install it before upgrading. - !The
VC-WIN64I(Windows-on-Itanium) andVC-CE(Windows CE) build targets are dropped from Configurations.
Dreadnode adds white-box AI attack support — GCG-family, embedding, generator, and activation methods run against a local open-weight surrogate — plus three new jailbreak algorithms and two classifier-evasion attacks, letting red teams test model internals rather than just black-box prompts. A deterministic tool_call_attack_landed scorer gives a judge-free success signal for agentic attacks.
Dreadnode added a tree-search jailbreak attack tool for AI red-teaming, enriched findings exports with audit and compliance data, and introduced session pinning alongside more resilient hosted inference streaming.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
what's the number?
Each feature carries 0–100 for how completely the vendor documented it — not how big or important the work is. A major capability described in eight words scores low, and that is the finding.
- depth0–40 what it does and how it works or what changed
- specificity0–30 names real surfaces — APIs, flags, formats, limits, numbers
- actionability0–30 enough to go use it — a named endpoint, flag, or config key tops this; a UI path is a starting point
Code and config rank first by construction: a documented endpoint, flag or config key scores at the top of specificity and actionability, so it sorts above a feature described only in prose. Under 60 the notes go thin — everything below the “thinner coverage” line is thinner documentation, not smaller work. Hover any meter for that feature's three sub-scores.
Adds SLIP to the AI red-teaming attack tool list, performing tree-search-based prompt jailbreaks against a target model with configurable depth, branching, and judge models.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄ Launched from the red-teaming UI by selecting a project, choosing SLIP, setting the target model, and configuring depth, branching, and judge models before running the campaign.︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄
Sessions can now be pinned from the ... menu in the Chat sidebar or via Agents › Sessions. Pinned sessions appear at the top of the list and are visible only to the user who pinned them.
The findings parquet export gains new columns for audit trail, ASR, compliance tags, adversarial-ML metrics, and project-level rollups, and now reflects the latest operator-reviewed severity and status.
Hosted inference SDK streams dropped mid-connection are now retried up to 3 times with exponential backoff (1s, 2s, 4s) to reduce failures on unstable network paths.
- New detection New structural and long-context jailbreak attacks in the airt SDK
- New detection New classifier-evasion attacks in the airt SDK
Where each pick stands · Oct 23–29
Tools with changes in each pick's category over the last 12 weeks, including the pick itself.
Each pick's release rhythm · Oct 23–29
Changes per week for the last 12 weeks. The bright bar is this week; a dash marks a week with no daily.
Which categories moved · Oct 23–29
The thick bar is this week. The thin bar is the average of the four weeks before, counting only weeks with a daily.
Who shipped the most · Oct 23–29
Changes per tool this week. Volume is not a quality score.