BETA · GAMMA-Ray is published for review. The rules, the mapping and every rating here are experimental and will change. How a rating is made · what exists · send a correction
GAMMA-Rayv0.0.155 · beta

AISLE

GAMMA-Ray report · Spectrum 0.2 beta · Public tier · read 2026-11-10 · not reviewed by the company yet

AISLE sells a software platform that uses AI agents to find vulnerabilities in source code, triage them, write and test patches, and deliver them as pull requests for people to merge.

1solution carded
of 6 found
20sources captured
and hashed
68provisions lit
in 8 of 8 instruments
6examples of the agent
at work, with sources

Built only from AISLE's own public pages (product, blog, about, customer stories), third-party vulnerability records (OpenSSL, curl, Samba) and, for company basics only, BetaKit. AISLE has not seen this report. Nothing was tested and no credentials were used. If you are AISLE and something here is wrong or missing, the correction will be made and dated, and you can claim the card to add evidence.

The agentic solutions

One card per solution. Each level is computed from the evidence by a published rule; a dimension with nothing public reads "not yet evidenced", which is not a no. A card rates a product as documented, not anyone's deployment of it.

AISLE platform (AI-native Cyber Reasoning System): Vulnerability Discovery, Vulnerability Triage, Autonomous Remediation, Continuous Improvement

Analyzes connected repositories, verifies and ranks findings, writes patches, tests them in an isolated container and in the customer's CI, and opens pull requests that people merge.

AISLE platforma product, as documented
SPECTRUM 0.2 BETA · Public tier · 2026-11-10
A4AutonomyOrchestrateswith options on · as shipped not yet evidenced
π Moderate · S1·R5·A1 ×3
O2OpennessDocumented
π Moderate · S1·R5·A1 ×3
DraftStage: Building (9 of 16)In production

Standout. Autonomy 4 (Orchestrates): "AISLE plugs into the scanners, repos, and CI you already run"

Next unlock. Control 2 (Configurable): public evidence that what it can reach is enumerated: scopes, permissions, tools would get there.

Trust. FACT-Ray: 17 facts: 17 π Moderate · found by an agent, captured and hashed, checked by 2 other agents (same model family, shared repository); no human has signed it.

Standards constellation. 68 provisions lit across 8 of 8 instruments.

Evidence: 11 sources, captured and hashed · the card's page has the vault's read key · Not reviewed by its vendor yet · correct or claim this card

The full card, every answer and its evidence · the governance layer

AISLE Snapshot

A one-time AI code audit that delivers prioritized findings, deployable inside the customer's perimeter. No card yet (available): this solution has not been assessed.

Dependency Firewall (AI package firewall)

A read-only proxy for crates.io, npm and PyPI that checks package requests and blocks malicious packages. No card yet (available (rolled out October 2026)): this solution has not been assessed.

AISLE MCP Server

A read-only MCP server with five tools that lets coding agents such as Claude Code, Codex and Cursor read AISLE findings and fixes. No card yet (announced 9 November 2026; read-only): this solution has not been assessed.

Proof of exploitability for web apps and APIs

With customer authorization, an agent sends targeted requests to a running test application to confirm a finding and attaches the evidence. No card yet (rolled out October 2026 (blog)): this solution has not been assessed.

AISLE analyzer for open source

No-cost use of the analyzer for eligible critical open-source projects, reviewing pull requests (e.g. OpenSSL, curl, OpenEMR). No card yet (available on application): this solution has not been assessed.

Our reading

The shape. The AISLE platform reads A4 · R2 · C1 · O2 (Building). It builds a customer's repository in an isolated container, writes a patch, runs the customer's tests and opens a pull request; it plugs into scanners, repositories and CI. A merge to the main branch always waits for a person on the customer's team.

Where the evidence is strongest. The work is visible outside AISLE's own pages. OpenSSL's vulnerabilities page, curl's machine-readable vulnerability list and Samba's advisory credit AISLE's research team, and AISLE is a CVE Numbering Authority. The card counts what the agent does on its own from a named customer: Productboard rolled AISLE out across all 56 repositories, with automated triage and AI-generated pull requests.

Autonomy 4, as an option. How far AISLE goes is a setting the customer chooses per severity and per repository: stop at triage, or go on to a tested pull request. The page states no default, so Autonomy 4 is marked available.

Control 1, and what would lift it. The merge gate is documented. Level 2 asks for access narrower than the whole account, stated for AISLE itself: a public page on the scopes AISLE asks for when it connects to GitHub, GitLab, Bitbucket or Azure DevOps would meet it. The read-only MCP server limits what other agents can read from AISLE, which is a different thing, and is shown, not counted.

Openness. Data use and the limits of what it fixes are documented, which meets Disclosed and Documented under Spectrum 0.2. Published evaluation results with a method would reach level 3; the headline figures on its pages have none.

For readers. On 15 October 2026 AISLE and Cohere signed a letter of intent for Cohere to acquire AISLE; nothing we read says it has completed.

What would move it

The one piece of public evidence that would lift each card next:

The agent at work

Examples from public pages that name a customer or give figures. Each source was captured and hashed on the date shown.

OpenSSL Foundation

AISLE's story says the Foundation uses AISLE internally to analyze pull requests and generate fixes, and that AISLE found 20 of 23 OpenSSL CVEs in six months; OpenSSL's own vulnerabilities page credits Aisle Research staff as finders and fix developers. "Now AISLE is analyzing pull requests (PRs), flagging issues, and generating fixes"

Source (captured 2026-11-10T15:46Z, sha256 eb1462934ed8c67f)

OpenSSL (third-party record)

OpenSSL's vulnerabilities page lists Aisle Research staff as finders of multiple CVEs and as fix developers. The credit names a person of AISLE's research team; it shows the team's work, which the card does not count as the agent acting on its own. "Joshua Rogers (Aisle Research)"

Source (captured 2026-11-10T06:54Z, sha256 b8f6515103b07f6f)

curl project

AISLE's story says curl has used AISLE to find and propose fixes since February 2026, with 35 CVEs attributed to AISLE in 12 months; curl's vuln.json credits Aisle Research as finder on 34 CVE entries. "bringing AISLE’s 12-month total to 35 CVEs"

Source (captured 2026-11-10T06:54Z, sha256 a9b30d11d7312a01)

curl (third-party record)

curl's machine-readable vulnerability list credits Aisle Research staff as finders, and in some entries as remediation developers. The credit names a person of AISLE's research team; it shows the team's work, which the card does not count as the agent acting on its own. "Joshua Rogers (Aisle Research)"

Source (captured 2026-11-10T06:54Z, sha256 97e81f9a06590abd)

Productboard

Customer story: rolled out across all 56 repositories, reducing 30,000 alerts to six actionable items, with Jira, Slack and AI-generated pull requests. "Productboard rolled out AISLE across all 56 repositories"

Source (captured 2026-11-10T15:46Z, sha256 b90ab4d495f89124)

Samba (CVE-2025-10230)

Samba's advisory for a CVSS 10.0 WINS hook command injection credits a researcher of Aisle Research. The credit names a person of AISLE's research team; it shows the team's work, which the card does not count as the agent acting on its own. "Reported by Igor Morgenstern of Aisle Research."

Source (captured 2026-11-10T06:54Z, sha256 5ba8380d2c449c67)

The company

The people

Public and professional only, from the company's own pages or a public professional source.

The graph

The company, its people and the stories on the left; each solution, the captured sources behind its card, the indicators those sources evidence and the provisions of the instruments those indicators touch, to the right. "Touched" never means "complies".

people and storiescompanysolutions11 sources17 indicators68 provisions, by instrumentI04: The remediation FAQ states AISLE builds the repository in an isolated container, applies its patch and runs the tests.I08: The page states AISLE writes a patch, runs it against the customer's tests and opens a pull request.I10: The platform page states it plugs into the customer's scanners, repositories and CI.I11: AISLE connects from its own service to the customer's hosted version control systems.I18: The page states AISLE writes and tests a patch and opens a pull request, that the customer decides per severity and repository how far it goes, and that a person on the customer's team merges; no default is stated.I20: The home page states what the system is for: finding, fixing and verifying vulnerabilities in source code.I21: The customer sets how far AISLE acts on its own, per severity and per repository.I22: The page states which findings get guidance instead of a patch, and that a fix breaking tests goes back to the team.I26: AISLE's OpenSSL story states maintainers invoke it by tagging the GitHub handle @aisle-analyzer. Stated by AISLE; the account was not checked on GitHub.I38: A merge to the main branch needs a person on the customer's team; AISLE opens the pull request and waits.I41: The enterprise FAQ states every finding and fix is logged through to the approved pull request, with timestamps and approvers; documented only, not observed.I57: Users can contest a triage decision by explaining why and sending the issue back for re-assessment.I58: The privacy policy lists customer data as code repositories and related data submitted to the service.I67: The enterprise page states a figure for engineering hours saved, and the remediation page gives monthly patch and hours figures; no method is published.I68: The deployment comparison states updates and intel are continuous for Managed SaaS and delivered monthly for private cloud and on-prem; this is an update cadence, with no improvement process described.S1: A customer story names Productboard, which rolled AISLE out across all 56 repositories and integrated its AI-powered pull requests into its workflow; the named customer is what is counted, as the page publishes no count of pull requests or tasks. CVE credits to Aisle Research name staff researchers and are not counted here.S4: AISLE's announcement post, published 9 November 2026, states its MCP server gives a customer's coding agent, such as Claude Code, OpenAI Codex or Cursor, read-only access to AISLE findings through five tools, with sign-in by OAuth or a personal access token.AISLEOndrej VlcekJaya BalooStanislav FortOpenSSL FoundationOpenSSL (third-party record)curl projectcurl (third-party record)ProductboardSamba (CVE-2025-10230)AISLE platform (AI-native Cyber Reasoning System): Vulnerability Discovery, Vulnerability Triage, Autonomous Remediation, Continuous ImprovementAISLE platformA4 · R2 · C1 · O2AISLE SnapshotAISLE SnapshotDependency Firewall (AI package firewall)Dependency FirewallAISLE MCP ServerAISLE MCP ServerProof of exploitability for web apps and APIsProof of exploitabilityAISLE analyzer for open sourceOpen-source analyzeraisle.com/platform/autonomous-rem…aisle.com/platformaisle.com/platform/vulnerability-…aisle.comaisle.com/customers/opensslaisle.com/enterpriseaisle.com/platform/vulnerability-…aisle.com/privacy-policyaisle.com/deployment/managed-saasaisle.com/customers/productboardaisle.com/blog/announcing-the-ais…It generates and executes code with local or cloud privilegesI04It executes actions itselfI08It works across several domains or systemsI10It connects to external tools or services across a trust boundaryI11The agent works independently; the person approves critical decisions onlyI18Its purpose and permitted tasks are stated in writingI20Its autonomy limits and the tools it may use are definedI21Its limits and known failure cases are documented for those who use itI22It has its own identity, separate from any person'sI26High-impact actions need a person's approvalI38Its actions are loggedI41There is a channel to report problems with it or contest its outputI57The data it uses is describedI58Its value is measuredI67It is improved on a regular cycleI68It is in production: a public source names a customer or gives usage figures for what it does on its ownS1It exposes an agent interface: a published MCP server, an agent card, or an API documented for agents to callS4Salesforce 18EAAMM 2OWASP 9MCP authorization 1AAGMM 6Microsoft 13NIST AI RMF 16EU AI Act 3OWASP, AT4 Code Executing AgentSalesforce, Level 2: Simple Orchestration, Single DomainSalesforce, Level 1, how to advance from Level 1 to Level 2Microsoft, 200: Repeatable, Business strategySalesforce, Level 3: Complex Orchestration, Multiple DomainSalesforce, Level 2, how to advance from Level 2 to Level 3Salesforce, Level 3, how to advance from Level 3 to Level 4Microsoft, 400: Capable, Business strategyMicrosoft, 200: Repeatable, Technology and dataOWASP, AT6 Externally Extended AgentNIST AI RMF, GOVERN 6.1NIST AI RMF, MAP 4.1EAAMM, L4 Supervisory (survey item Q9)NIST AI RMF, MAP 1.1NIST AI RMF, MAP 2.1NIST AI RMF, MAP 3.3NIST AI RMF, MAP 1.3EU AI Act, Article 13OWASP, Maturity Level 1 Experimentation without Guardrails, key enterprise actionNIST AI RMF, MAP 2.2Microsoft, 300: Defined, Business strategyNIST AI RMF, MEASURE 2.5AAGMM, GD-02Salesforce, Level 2, how to advance from Level 2 to Level 3OWASP, Maturity Level 4 Adaptive, Self-Regulating Governance, key enterprise actionOWASP, Maturity Level 2 Policy-Defined, Human-in-the-Loop, key enterprise actionEU AI Act, Article 14AAGMM, GD-06NIST AI RMF, MAP 3.5Salesforce, Level 3, how to advance from Level 3 to Level 4OWASP, Maturity Level 0 Unaware and Ad Hoc, key enterprise actionOWASP, Maturity Level 1 Experimentation without Guardrails, key enterprise actionEU AI Act, Article 12AAGMM, GD-03AAGMM, GD-08NIST AI RMF, MEASURE 3.3NIST AI RMF, GOVERN 5.1Salesforce, Level 1, how to advance from Level 1 to Level 2Salesforce, Level 4, how to maximize valueNIST AI RMF, GOVERN 5.2AAGMM, GD-05EAAMM, Readiness dimension 2Salesforce, Level 0, how companies can advance from Level 0 to Level 1Salesforce, Level 1, how to advance from Level 1 to Level 2Salesforce, Level 0, how companies can advance from Level 0 to Level 1Salesforce, Level 0, how companies can advance from Level 0 to Level 1Salesforce, Level 1, how to advance from Level 1 to Level 2Salesforce, Level 3, how to advance from Level 3 to Level 4Salesforce, Level 4, how to maximize valueSalesforce, Level 4, how to maximize valueMicrosoft, 200: Repeatable, Business strategyMicrosoft, 300: Defined, AI strategy and experienceMicrosoft, 300: Defined, Business strategyMicrosoft, 400: Capable, Business strategyMicrosoft, 400: Capable, AI strategy and experienceNIST AI RMF, MAP 1.4NIST AI RMF, MAP 3.1Microsoft, 100: Initial, Organization and cultureAAGMM, GD-12Microsoft, 400: Capable, AI governance and securityMicrosoft, 500: Efficient, Business strategyMicrosoft, 500: Efficient, AI strategy and experienceNIST AI RMF, MANAGE 4.2OWASP, Maturity Level 1 Experimentation without Guardrails, key enterprise actionNIST AI RMF, MEASURE 4.3MCP authorization, Authorization (draft)OWASP, Maturity Level 1 Experimentation without Guardrails, key enterprise actionSalesforce, Level 3, how to advance from Level 3 to Level 4

The evidence

SourceReadSHA-256 (first 16)
aisle.com/2026-11-10T15:46Zb44da331375d21e2
aisle.com/about-us2026-11-10T06:54Zc2435d6bd46472c3
aisle.com/blog/aisle-and-cohere-to-join-forces2026-11-10T06:54Z3d719f8785c2bfd6
aisle.com/blog/announcing-the-aisle-mcp-server2026-11-10T15:46Z8fba9a52b1dcaf4c
aisle.com/customers/curl2026-11-10T06:54Za9b30d11d7312a01
aisle.com/customers/openssl2026-11-10T15:46Zeb1462934ed8c67f
aisle.com/customers/productboard2026-11-10T15:46Zb90ab4d495f89124
aisle.com/deployment/managed-saas2026-11-10T15:46Z3a90cd5d7867f136
aisle.com/enterprise2026-11-10T15:46Zdff44af85ee36354
aisle.com/newsroom/press-releases/aisle-achieves-expanded-scope-as-cve2026-11-10T06:54Za80b1b9e5262bce7
aisle.com/newsroom/press-releases/aisle-emerges-from-stealth2026-11-10T06:54Z20d872d1ee49c60e
aisle.com/platform2026-11-10T15:46Z9ede4e7781a0a527
aisle.com/platform/autonomous-remediation2026-11-10T15:46Z781c682dd9303d6b
aisle.com/platform/vulnerability-discovery2026-11-10T15:46Z56f5b867d7f837df
aisle.com/platform/vulnerability-triage2026-11-10T15:46Z2c835b20038c8e24
aisle.com/privacy-policy2026-11-10T15:46Z13cbb99908547db4
betakit.com/cohere-to-acquire-czech-ai-cybersecurity-firm-aisle-set-up2026-11-10T06:54Zd4e66a70222918a2
curl.se/docs/vuln.json2026-11-10T06:54Z97e81f9a06590abd
openssl-library.org/news/vulnerabilities/index.html2026-11-10T06:54Zb8f6515103b07f6f
www.samba.org/samba/security/CVE-2025-10230.html2026-11-10T06:54Z5ba8380d2c449c67

How this report was made

A researcher-assessor gathered and captured the sources and mapped the evidence to indicators, without assigning a level. A challenger argued against every level three and four and every doubtful credit. The deterministic scorer computed the levels and the gate checked the rules. A card appears here only after two rounds of review and a sign-off by agents that did not write it (the team's workflow); AISLE platform (AI-native Cyber Reasoning System): Vulnerability Discovery, Vulnerability Triage, Autonomous Remediation, Continuous Improvement: reviewed in 3 rounds, signed off by gamma.approve@web-summit.dinis on 2026-11-10. A person's sign-off of the series is still to come.

RoleWho
researcher-assessora sub-agent of gamma.studio@web-summit.dinis
challengera sub-agent
scorerbuild/spectrum.mjs
analyst, designer, writergamma.studio@web-summit.dinis
reviewgamma.review@web-summit.dinis, 3 rounds
sign-offgamma.approve@web-summit.dinis, 2026-11-10

The vault

Every capture, the dossier, both assessors' mappings, the challenger's notes and this report's data are in a public sgit vault (z11yqwik). Read key:

sgit_public_read_5dcfc650a15adce6282cb5996d1cd2f57f8bd817588886e419e57a94246fdabd:z11yqwik

With sgit: sgit clone followed by the key. The key reads; it cannot write.

Corrections

None yet. A correction is made in place, dated here, and carried to the card, the vault and the article.

The article

AISLE on the GAMMA-Ray Spectrum: A4 · R2 · C1 · O2, written from this report for PrompTugal.news.