Kenvara
Cloud security statement
How Kenvara is built, hosted and secured, and how it decides which customers see which known issues.
Security contact: support@itsm-ltd.com, with “security” in the subject.
1. Purpose and summary
This statement describes the security of Kenvara (the “App”), published on the Atlassian Marketplace by ITSM Ltd (company number 17339600). It is written to support your security review and to be shared with your risk, procurement and information security teams.
The single most important fact about this App’s hosting is that it holds no data outside Atlassian. The App is built entirely on Atlassian Forge. We operate no servers, no databases and no hosting infrastructure for it, and it makes no outbound calls to systems we control. Most of the questions a security review would ask about a vendor’s hosting environment are therefore answered by Atlassian’s own controls, not ours.
Kenvara is a known error database for Jira Service Management. It keeps known errors in its own storage on your site and shows selected ones on your customer portal, as known issues, to the customer organisations entitled to see them. One property defines it: no portal response ever contains a known error the requesting user’s organisation is not entitled to see.
At a glance
| Question | Answer |
|---|---|
| Where does the App run? | Atlassian Forge only |
| Where is customer data stored? | Forge SQL and Forge’s key-value store, inside the customer’s installation |
| Does data leave Atlassian’s infrastructure? | Kenvara declares no egress. Only Jira’s own notifications and an export an administrator takes leave the site |
| Do you host any part of the App? | No |
| Which Atlassian products? | Jira Service Management: the customer portal, the agent views, and project and site settings. Not Customer Service Management |
| Does the App write to Jira? | Yes, in five ways (section 5.2) |
| Does the App run background jobs? | Yes: daily checks, an hourly licence check, product events, queued work, and a switch-off on uninstall (section 5.2) |
| Can your staff read customer data? | No. We read Kenvara’s log lines, which hold no account id or record text (section 6.1) |
| Is data encrypted? | Yes, in transit and at rest, by the Atlassian platform (section 3.2) |
| Is data residency supported? | Yes, inherited from the site; not on Atlassian Government Cloud (section 3.4) |
| Does the App use AI or machine learning? | No, not in this version (section 4.3) |
| Does the App send analytics or telemetry? | No. Counts and feedback stay in the customer’s site (section 4.2) |
| Are you SOC 2 or ISO 27001 certified? | No; Atlassian’s infrastructure is (section 9) |
| Sub-processors | Atlassian; Google Workspace for support email only (section 10) |
2. Architecture and hosting
2.1 Kenvara is a Forge app on the nodejs22.x runtime. Its pages are Forge UI Kit modules inside
Jira, except the export and import page, a Forge Custom UI page served by Atlassian from Kenvara’s own
bundle.
2.2 We do not provision, operate, patch or monitor any infrastructure for the App. Operating system patching, runtime patching, network security, physical security, capacity and availability of the underlying platform are Atlassian’s responsibility.
2.3 The App contains no Atlassian Connect modules, no Forge remotes and no externally hosted components.
2.4 Shared responsibility. Atlassian secures the platform. We are responsible for the App’s own code, its declared scopes, its data handling logic, its dependencies and its release process. You are responsible for administering user access within your Atlassian site, for the roles your people hold, for the membership of your customer organisations, for the audiences you publish known issues to, for the content your users place in the App, and for deciding whether the App is appropriate for your data.
3. Data storage, encryption, residency and retention
3.1 Storage. Kenvara stores its records in Forge SQL, in 13 tables. It keeps short-lived caches, counters, flags and one secret in Forge’s key-value store, never record data or rich text. Atlassian isolates both per installation. Kenvara’s code also keeps nothing in memory between calls, because Forge reuses warm containers across sites. The App has no other data store.
3.2 Encryption. Encryption is provided by Atlassian, and we do not implement or override it. Atlassian states, on its Security Practices page, read on 20 October 2026: “Any customer data in Atlassian cloud products is encrypted in transit over public networks using TLS 1.2+ with Perfect Forward Secrecy (PFS) to protect it from unauthorized disclosure or modification”, and “Data drives on servers holding customer data and attachments in … Jira Service Management Cloud … use full disk, industry-standard AES-256 encryption at rest.”
3.3 Secrets. Kenvara holds one secret: a random key, created for each installation and kept in Forge’s secret storage. It is used only to turn a portal customer’s account id into the one-way code that feedback is kept under. Kenvara uses no credentials, API keys or tokens, and none is in its source code, its manifest or its repository.
3.4 Data residency: what is in scope and what is not. Forge SQL and Forge’s key-value store follow the data residency of the Jira site the App is installed on. Where you have pinned Jira to an Atlassian region, Kenvara’s data is pinned to the same region and moves with your product data if you relocate. Data residency is administered by Atlassian, which publishes the current list of regions. Atlassian asks each Marketplace Partner to publish what its app treats as in scope and out of scope. For Kenvara:
In scope: everything Kenvara stores, in Forge SQL and Forge’s key-value store. That is known errors and their versions, audiences, links, applied workarounds, the history of changes, settings, export and import working data, caches, and portal counts and feedback.
Out of scope, because Kenvara does not store it:
- the Jira and Jira Service Management data it reads or writes, which follows Jira’s own residency;
- Jira’s notifications;
- export files your administrators take;
- Forge’s log lines, which Atlassian keeps;
- our support correspondence (section 10).
Kenvara is not available on Atlassian Government Cloud, where Forge SQL is not offered.
3.5 Backup and recovery. Atlassian backs up Forge storage. Data Kenvara deletes can be restored only by Atlassian, on a support request within 21 days, so deletion is final from the app’s point of view. We hold no backup of your data: your export file is your backup. Recovery objectives are those of the Atlassian platform; we do not offer separate RTO or RPO commitments.
3.6 Retention, and what cannot be deleted. Kenvara keeps known errors, their versions and the history of changes for the life of the installation. It deletes per-customer feedback after 30 days, and working copies of exports and imports within about two days of the job ending. Section 10 of the privacy policy gives every period.
The history of changes is append-only: the only change ever made to it replaces a closed account’s id with “erased”. That guarantee rests on Kenvara’s code, not on the storage platform. One function is the only route to that change, and two independent tests, among the checks in section 7.2, fail on any delete from the history’s table, or on any update made outside that function. Known errors cannot be deleted either: they are withdrawn or retired, which takes them off the portal and keeps their history. The implications under data protection law are set out in section 12.1 of the privacy policy and clause 9 of the Data Processing Agreement.
3.7 Deletion. Uninstalling Kenvara offers a choice. If you keep access, Atlassian keeps Kenvara running until the subscription or evaluation ends, plus 60 days. If you remove it immediately, it goes at once and any time left is given up. Either way, Atlassian then deletes its known errors under its own retention rules, up to about 30 days after it goes: its storage documentation gives 28 days, and its data-lifecycle page refers to the retention period in Atlassian’s SOC 2 report. We retain no copy. If you reinstall and ask us within 21 days of uninstalling, we can ask Atlassian to relink your earlier data with your consent; after that it cannot be recovered.
3.8 Growth. Kenvara never prunes the history of changes, even near Forge SQL’s limit of 1 GiB per installation. No customer data is discarded without notice.
4. Data egress
4.1 The App’s manifest declares no external egress: no permissions.external block of any kind,
and no Forge remotes. The Forge platform blocks outbound traffic to any domain not declared, and
Atlassian reviews declared egress at app approval. Because Kenvara declares none, it cannot send customer
data outside Atlassian’s infrastructure. Its code calls only Jira and Jira Service Management, and
Atlassian’s own licence and personal data reporting services. Kenvara’s deployment command checks that
this is still so before every deployment it makes (section 7.2).
4.2 No analytics, telemetry service or error reporting. Kenvara uses no analytics service, no third-party tracking and no error reporter, and sends nothing about your use outside Atlassian. The only record of use we see is Kenvara’s log lines (section 6.1.2), which identify the site but no person. While its licence is active, it counts, for each known issue, how often it was shown and opened and how readers answered. The counts identify no one, and are kept on your site. Each answer is also kept for 30 days with a one-way code in place of who gave it, so that repeated presses by one person are counted once; after that only counts remain. An administrator can delete every customer’s answers at once.
4.3 No artificial intelligence or machine learning. This version of Kenvara contains no AI or machine learning feature. It makes no call to any large language model or inference service, whether Atlassian’s or a third party’s. Customer data is not used to develop, train, fine-tune or evaluate any model, by us or by anyone else; clause 4.4 of the Data Processing Agreement makes this a contractual undertaking. A version that adds Forge’s AI feature needs your administrator’s approval, and before any AI feature is added, this statement will describe it.
4.4 Rich text and media. Known-error text is stored in Atlassian’s document format and cleaned before it is saved for customers. Images and other media, mentions (which would carry an account id), cards and extensions are removed, and the portal cleans the text again before every answer. A customer-facing field that still holds an image cannot be published.
5. Permissions and least privilege
5.1 The App requests the ten scopes below and no others. Where a classic scope grants nothing beyond
the reads and writes Kenvara makes, it is classic; where the classic scope would add power Kenvara never
uses, it is the granular one. Kenvara declares no scope it does not use, never manage:jira-configuration,
and nothing that would put a consent screen in front of a portal customer.
| Scope | Kind | Why Kenvara needs it |
|---|---|---|
storage:app |
Forge | Forge SQL and the key-value store: the records and the caches |
read:jira-work |
Classic | Reading work items, permissions and project configuration, as the agent; background status reads |
write:jira-work |
Classic | Raising and linking a problem, and the review reminder |
read:servicedesk-request |
Classic | Service project, request type and request reads |
write:servicedesk-request |
Classic | An agent’s public reply with a workaround; Jira Service Management checks the agent’s right to post it |
read:organization:jira-service-management |
Granular | Customer organisations, for membership; the classic scope would also let Kenvara create and delete customers |
read:servicedesk.organization:jira-service-management |
Granular | A service project’s organisations, for the publish-time check and Preview as organisation |
read:servicedesk:jira-service-management |
Granular | The event that a service project was deleted, so its known errors leave the portal |
write:project.property:jira |
Granular | The one project property that tells Jira where to show Kenvara’s panel |
report:personal-data |
Forge | Reporting the account ids Kenvara holds, and erasing closed accounts |
5.2 How the App calls Jira. Kenvara acts as the person using it wherever it can, so it sees only what they could. It acts as the app in seven named places, each held by a test, among the checks in section 7.2, that fails on any other:
- the organisations of customers who sign in with an Atlassian account, whom Atlassian does not let an app act as;
- its own licence;
- linked work items’ status, for retirements, reconciliation and imports, and the site’s organisations once a day;
- the review reminder;
- a check that a publisher holds the publishing role, which can only refuse;
- personal data reporting;
- a member-list confirmation, switched off in this version.
It writes to Jira in the five ways the privacy policy’s section 5.2 lists, and deletes nothing there. It also runs without a person: daily checks, an hourly licence check, responses to product events, queued exports and imports, and a switch-off when it is uninstalled.
5.3 The App does not request, collect or store Atlassian passwords, API tokens or Personal Access Tokens. It uses no credentials of any kind.
5.4 Any change that adds a scope requires your site administrator’s explicit approval before the new version is installed.
5.5 Content Security Policy. Kenvara relaxes no part of Forge’s content security policy. Its one Custom
UI page, for export and import, uses no inline script or style, builds every element from text, never uses
innerHTML, and downloads the export file from memory.
6. Access control: yours, and ours
6.1 Our access to your data
6.1.1 We have no means of reaching your data. Kenvara has no back door, no support console and no export to us, and we never ask for your export file.
6.1.2 We can read Kenvara’s log lines in our developer console. A line holds identifiers, codes and counts. Kenvara’s logging code drops any line with a field for an account id, an email address, a display name, an organisation or record text, and any value shaped like an account id. Every portal call writes one line, and errors write their own. Atlassian keeps them for 30 days, with the site each came from; a site administrator can switch off developer access to the app’s logs. We use them for a daily review of security signals, and to count how many sites use Kenvara each week.
6.1.3 The only other way we see your data is if you send it to us, for example a screenshot attached to a support email, or the publication history report your administrator can create in an incident, which holds no known-error text and no account ids. We ask customers not to send more than is needed, and handle such material in accordance with our privacy policy.
6.1.4 ITSM Ltd is one person. Access to the developer space is held by Steven Godson alone, protected by multi-factor authentication, and its membership is reviewed every month. The AI coding agents he uses work under his direction, and never deploy to production. Anyone ITSM Ltd later engages will be bound by written confidentiality obligations and will receive security awareness training before being given access.
6.2 How the App decides who sees what
This matters more for Kenvara than for most apps, because a known error shown to the wrong customer organisation is a breach between one customer’s clients. We describe the model in full.
6.2.1 Identity. Identity comes only from the context Atlassian signs into each call, never from the page, the request or the browser.
6.2.2 Three server functions. Kenvara has three, one each for the portal, agents and settings, each with its own checks. The portal function writes nothing a customer could use to change a known error.
6.2.3 Agents and administrators. Agents are checked for an agent licence and the service project’s agent permission on every call; a confirmed answer is reused for five minutes for reading only, and every write checks again. Administrators are checked live, first, every time.
6.2.4 The portal. A known issue is shown only when every condition holds, and anything Kenvara cannot work out hides it: a failed lookup, an unknown kind of caller or a missing value. Membership of a customer organisation is read from Jira Service Management on each call, or from a cache your administrator sets between 0 and 300 seconds (60 by default). When someone is removed from an organisation, they stop seeing its known issues within 2 minutes at Kenvara’s default setting. Jira may take longer to report a removal. A page already open keeps what it showed until it is reloaded. When a removal must take effect at once, unpublish the known issue or change its audience first.
6.2.5 What is never used to decide. Kenvara never uses a request’s Organizations field, its participants or request-sharing settings. The portal is never told who else can see a known issue, how many were hidden, or why one was not shown: “not linked”, “not found” and “not entitled” answer identically. The licence can only ever remove what customers see, never add to it.
6.2.6 Customers who sign in with an Atlassian account. Section 9.5 of the end user terms sets out the limits that apply to them in this version, and the automatic and manual switch-offs.
6.2.7 The honest architectural statement. Forge has no equivalent of row-level security in a database. App storage belongs to the App, so every read and write runs with the App’s authority, and the separation between what one customer organisation may see and what another may see is enforced in Kenvara’s server code. We say this plainly because a statement claiming that the storage layer enforces it would be false. What makes the boundary trustworthy is the model above, and the adversarial leakage suite in section 7, which blocks every release.
6.2.8 The interface predicts the boundary; it is not the boundary. Hiding a control from someone who may not use it is good manners. The refusal happens on the server.
7. Secure development
7.1 Release-blocking tests. An adversarial suite of cross-organisation leakage tests blocks every release: any test that fails means no release, and only Steven Godson can waive one, on the record.
7.2 Automated checks. Kenvara’s deployment command runs these checks, among others, before every deployment it makes, and deploys nothing unless all of them pass:
- type checking (TypeScript strict) and linting;
- the manifest’s rules, and a check that the App declares no egress;
- source sweeps: parameterised SQL only, app-level calls only in the named files, no state kept between calls;
- the unit and property tests, and the leakage suite’s tests under both licence states;
- the documentation checks;
- a dependency audit, failing on high and critical advisories, with dated and reviewed exceptions.
Continuous integration is separate from the deployment command. On every pull request into the main branch, it runs gitleaks, a secret scanner, over the full history of the code the pull request would merge and over its files; the gitleaks download is checked against a pinned checksum. When a pull request changes Kenvara’s website, it also builds the website from the website’s folder alone, installing its packages offline. Once a week, on the main branch, it runs Semgrep with Kenvara’s own rules, the dependency audit above and a check of package signatures. Apart from that weekly audit, it runs none of the checks listed above.
7.3 Dependencies. Kenvara has nine runtime dependencies: seven Atlassian Forge packages, React and zod. Each is pinned to an exact version, and the deployment command checks them against the lockfile (section 7.2). Kenvara adds no HTTP client, database driver or template engine of its own. Once a month, Dependabot proposes updates to the App’s dependencies, and it proposes a security update whenever an advisory affects a version Kenvara uses and a fixed version exists. Each is a pull request, reviewed before it is merged. Continuous integration scans it for secrets but does not run the App’s tests on it: the deployment command runs the checks in section 7.2 on it before any deployment it makes. The runtime is Node.js 22, the same in the manifest, the package manifest and continuous integration.
7.4 Input handling. Every request to Kenvara’s server functions is checked against a strict schema before it reaches the logic. Rich text is cleaned when it is saved, and again when the portal reads it.
7.5 Logging. Section 6.1.2 describes what Kenvara logs. Its logs are designed not to carry personal data, in line with Atlassian’s security requirements for cloud apps.
7.6 Environments and release. Development, staging and production are separate Forge environments. Only Steven Godson deploys production, from a clean, tagged commit whose built files match those the release checks ran on.
7.7 Kenvara’s listing and each of its versions are subject to Atlassian’s Marketplace review.
8. Vulnerability management and incident response
8.1 Reporting a vulnerability. Report a suspected vulnerability to support@itsm-ltd.com, with “security” in the subject. A report about what a customer could see is read before anything else. We acknowledge reports within 2 business days and will keep you informed of progress. We ask that you allow us a reasonable period to remediate before public disclosure, and we will not pursue researchers acting in good faith.
8.2 Remediation targets. We remediate confirmed vulnerabilities in accordance with Atlassian’s Security Bug Fix Policy for Marketplace apps, measured from report or triage:
| Severity (CVSS) | Target |
|---|---|
| Critical (9.0 to 10.0) | 10 days |
| High (7.0 to 8.9) | 4 weeks |
| Medium (4.0 to 6.9) | 12 weeks |
| Low (0.1 to 3.9) | 25 weeks |
These figures are Atlassian’s published timeframes for cloud apps, read on 20 October 2026. Atlassian revises its requirements periodically; where this statement describes an Atlassian policy, the Atlassian source is authoritative.
8.3 Incident notification to Atlassian. On a suspected leak between customer organisations, we raise a priority-1 ticket with Atlassian within 24 hours, under its partner incident programme, and update it at least every 6 hours until the leak is contained.
8.4 Incident notification to you. Where an incident affects your data, we tell your site’s administrators without undue delay, and we aim to do so within 72 hours of becoming aware. We reach them through the contacts on your licence, through Atlassian or in Kenvara itself, and give updates as the investigation progresses. Where we act as processor, we help you meet your own obligations; we do not notify the Information Commission on your behalf. Containment comes first: unpublishing a known issue, changing its audience, or switching off organisation-scoped known issues for customers who sign in with an Atlassian account.
8.5 Patch delivery. Because the App is Forge-only, minor and patch releases propagate automatically across installations without administrator action, so security fixes reach all customers shortly after we deploy them.
8.6 We have registered a security contact with Atlassian, as the Marketplace requires.
9. Certifications: an honest statement
We hold no independent security certification. ITSM Ltd is not certified to SOC 2, ISO/IEC 27001, ISO/IEC 27017, ISO/IEC 27018, PCI DSS or any comparable standard, and we do not claim to be.
What we can accurately say is this: the infrastructure on which the App runs is Atlassian’s, and Atlassian holds those certifications for its cloud platform. Because the App stores no customer data outside Atlassian’s infrastructure, the controls those certifications cover apply to the environment holding your data. You can verify Atlassian’s certifications and obtain reports directly from Atlassian at atlassian.com/trust. We are not able to supply Atlassian’s audit reports on Atlassian’s behalf.
9.1 Runs on Atlassian. Atlassian applies the Runs on Atlassian badge to apps that use only Atlassian-hosted compute and storage, declare no egress, and support data residency matching the host product. Kenvara meets those conditions, and its Marketplace listing carries the badge, which is Atlassian’s to apply.
10. Sub-processors
| Sub-processor | Purpose | Data involved | Location |
|---|---|---|---|
| Atlassian | Hosting, compute, storage; Marketplace licensing; Jira Service Management, our support tool | All of Kenvara’s data; and support correspondence held in our Jira Service Management | Per your data residency setting |
| Google Ireland Limited (Google Workspace) | Support email | Support correspondence only | Ireland / EEA |
Annex 3 of the Data Processing Agreement is the authoritative sub-processor list, giving the full legal entity and location for each; this table summarises it. We give 30 days’ notice of changes, under clause 7.2 of that agreement. No sub-processor other than Atlassian touches data held by the App.
11. Business continuity
11.1 Availability of the App depends on the Atlassian Cloud and the Forge platform, and continuity and disaster recovery for that platform are provided by Atlassian. Kenvara also depends on Atlassian’s licence service: if it cannot confirm Kenvara’s licence for more than 24 hours, customers see nothing from Kenvara until it can, and nothing leaks. We give no uptime commitment; see section 10 of the Support and Maintenance Description.
11.2 We commit to maintaining the App actively, including publishing at least one version update within every 18-month period, in line with Atlassian’s requirements for maintained Marketplace apps.
12. Compliance
- UK GDPR and Data Protection Act 2018: see the privacy policy at https://kenvara.itsm-ltd.com/legal/privacy-policy/ and the Data Processing Agreement at https://kenvara.itsm-ltd.com/legal/data-processing-agreement/.
- Information Commission registration: registered with the Information Commission, reference ZC207852.
- Atlassian Marketplace: we comply with the Marketplace Partner Agreement, the Atlassian Developer Terms and Atlassian’s security requirements for cloud apps. Kenvara reports the account ids it stores to Atlassian’s personal data reporting service every week, and erases those Atlassian reports closed.
13. Contact and updates
Security questions, questionnaires and vulnerability reports: support@itsm-ltd.com. We aim to respond to security questionnaires within 5 business days, consistent with the priority 4 target in the Support and Maintenance Description.
This statement is reviewed at least annually and whenever the App’s architecture materially changes. Where a change materially reduces the security commitments described here, we will give at least 30 days’ notice to the technical contact on your licence, and the change will not apply retrospectively or reduce our obligations during your then-current Subscription Term. Atlassian’s requirements, certifications and remediation timeframes change over time; where this statement describes an Atlassian control or policy, the Atlassian source is authoritative.
Published in accordance with the Atlassian Marketplace Partner Agreement. Read alongside the privacy policy, end user terms and support and maintenance description for Kenvara.