Mr.PlanB Logo
    Infrastructure Automation Ledger 2026
    Ledger
    Infrastructure Automation Ledger 2026

    Semaphore: Infrastructure Automation Ledger 2026 Scorecard

    A light web front end that schedules and runs Ansible, Terraform and OpenTofu, MIT with paid runners, SSO and HA, built mainly by one maintainer.

    Edition October 16, 20267 min read

    Where Semaphore UI put its effort, 2016 to 2026

    Each square is one capability area in one year. The brighter it is, the more the area gained in the Infrastructure Automation Ledger that year; 1, 2 and 3 mark that year's biggest pushes.

    ’16’17’18’19’20’21’22’23’24’25’26
    Provisioning breadth32
    Configuration management depth3
    State, planning and safety13
    Language and authoring
    Orchestration and self-service1
    Scale and control plane2
    Security and compliance1
    Bare metal and day-2 operations
    Ecosystem and integrations1
    AI assistance2
    no change+0.5+1+1.5 or moreScores out of 5 per area.

    Biggest pushes overall: State, planning and safety (+1.5), Orchestration and self-service (+1.5), Security and compliance (+1.5). Already strong in 2015: Orchestration and self-service, Security and compliance, Ecosystem and integrations.

    Top focus each year: 2023 Security and compliance; 2024 State, planning and safety; 2025 Ecosystem and integrations; 2026 Orchestration and self-service. 7 of 11 years show no gain in any area.

    Part of the Infrastructure Automation Ledger (edition 2026-10-16). How the scores work: methodology.

    Snapshot

    • Type: product (MIT Community edition plus Pro and Enterprise licences)
    • Owner: No legal entity named on the site; the licence file carries copyright for Denis Gukov (2021) and Castaway Labs LLC (2014 to 2021); Denis Gukov is the main author
    • Engine: None of its own; runs the Ansible, Terraform, OpenTofu, Terragrunt, shell, PowerShell and Python the operator installs
    • Licence: MIT for Community. Pro from $20 a month or $199 a year for 50 managed nodes and 500 managed Terraform or OpenTofu resources, one Pro user, one runner and one instance, 30-day trial; Enterprise by quote (https://semaphoreui.com/pricing, read 2026-10-15)
    • Latest release: 2.19.16 (2026-10-07); 2.18.31 patched the same day; 2.20.0-alpha4 (2026-10-12)
    • Support lifecycle: No published support policy; two stable lines are patched in parallel; Pro has a 48 business hour response SLA, Enterprise a custom one
    • Next milestone: 2.20 in alpha with LDAP and OpenID group mapping, no date; roadmap lists notifications and user-owned secrets for 2.21 and built-in inventory management for 2.22
    • Targets: Whatever the wrapped tool reaches: Linux, Windows and network devices through Ansible, clouds through Terraform or OpenTofu providers. No provider or module of its own for vSphere, Proxmox VE, Nutanix, Hyper-V, XCP-ng or KubeVirt; each is reachable only through Ansible collections or Terraform providers the operator installs
    • Tier: IaC management platform
    • Archetype: Founder-led orchestration layer, fast cadence
    • Evidence grade: B. Releases from 2024 onward were read from GitHub and the vendor's release page, pricing and feature tiers from the pricing page on 2026-10-15. The first release year, the legal entity and everything before 2024 are inferred, and the MCP documentation could not be fetched.

    Scope of the scores: Semaphore UI has no plan, dry run or drift feature of its own. It passes the inclusion test because it runs Ansible and Terraform, which do, so it is scored as a management platform for those engines and its provisioning and configuration scores reflect only what it adds on top of them. Scores before 2024 are held at the level of the early Ansible-only UI because no dated release history for those years was found.

    Scorecard

    DimensionScoreDirection
    Capability today41.3 / 100higher is better
    Momentum since end of 2022+21.7 points (2015 to 2022: +0.0)higher is better
    Credibility17.0 / 25higher is better
    Commercial risk9.0 / 25lower is better
    Technical lock-in3.0 / 25lower is better
    Integrator fit7.0 / 20higher is better
    AI leverage3.0 / 20higher is better
    Ledger Index53.3 / 100balanced weights; see the methodology

    Capability by domain, 2015 to today

    Scores 0 to 5 against fixed anchors; total is weighted to 100.

    Domain (weight)201520192022Today
    Provisioning breadth (15)1112
    Configuration management depth (12)1111.5
    State, planning and safety (12)0.50.50.52
    Language and authoring (10)1111.5
    Orchestration and self-service (10)2223.5
    Scale and control plane (8)1112.5
    Security and compliance (10)1.51.51.53
    Bare metal and day-2 operations (8)0000
    Ecosystem and integrations (8)1.51.51.53
    AI assistance (7)0001.5
    Total / 10019.619.619.641.3

    Year by year: 2015: 19.6 · 2016: 19.6 · 2017: 19.6 · 2018: 19.6 · 2019: 19.6 · 2020: 19.6 · 2021: 19.6 · 2022: 19.6 · 2023: 20.6 · 2024: 26.1 · 2025: 31.4 · 2026: 41.3

    What moved the score

    YearDomainChangePointsTrigger
    2023Security and compliance1.5 → 2+1.0OpenID Connect sign-in (vendor post 2023-07-23)
    2024Provisioning breadth1 → 1.5+1.52.11: Terraform and OpenTofu runs with an HTTP state backend
    2024Scale and control plane1 → 2+1.62.10.41: project runners (Pro)
    2024State, planning and safety0.5 → 1.5+2.42.11: Terraform and OpenTofu HTTP state backend (Pro)
    2025Configuration management depth1 → 1.5+1.22.13: tags and limits for Ansible runs from the template
    2025Ecosystem and integrations1.5 → 2.5+1.62.16: built-in Swagger API docs; webhooks and notification channels
    2025Provisioning breadth1.5 → 2+1.52.15: Terragrunt beside Terraform and OpenTofu
    2025Security and compliance2 → 2.5+1.02.12: two-factor authentication; 2.16: HashiCorp Vault integration
    2026AI assistance0 → 1.5+2.1Official MCP server listed in every edition without a GA or preview label; counted as a preview, half of the anchor
    2026Ecosystem and integrations2.5 → 3+0.82.17: Syslog; vendor-published Ansible collection and Backstage integration
    2026Language and authoring1 → 1.5+1.02.18: multiple variable groups; 2.14 template editor
    2026Orchestration and self-service2 → 3+2.02.17: one-time schedules, project export and import
    2026Orchestration and self-service3 → 3.5+1.02.19: workflows chain templates with an approval step; Docker and Kubernetes executors
    2026Scale and control plane2 → 2.5+0.82.19: Docker and Kubernetes executors; active-active HA in Enterprise (vendor post 2026-03-11)
    2026Security and compliance2.5 → 3+1.02.18: cloud secret managers and remote secret sync; 2.19: JWT identity tokens; audit UI only in the 2.20 alpha
    2026State, planning and safety1.5 → 2+1.22.19: workflows with an approval step that has a timeout and message

    Direction, 2023 to 2026

    Each release item carries one theme and one driver (V vision, C customer need, M market window, U upstream, P portfolio).

    DateItemThemeDriver
    2024-102.10.30: runners; multiple Vault passwords per templateScaleC
    2024-122.11: Terraform and OpenTofu HTTP state backendState and planningC
    2025-012.12: two-factor authenticationSecurityC
    2025-032.13: tags and limits for AnsibleConfiguration managementC
    2025-062.15: Terragrunt support and a visual redesignProvisioning breadthC
    2025-082.16: HashiCorp Vault integration and Swagger API docsSecurityC
    2026-022.17: one-time schedules, Syslog, project export and import; custom roles Enterprise onlyOrchestrationC
    2026-03Active-active high availability for EnterpriseScaleP
    2026-04Vendor comparison positions Semaphore UI as the replacement for the frozen AWXPackagingM
    2026-052.18: cloud secret managers, remote secret sync, multiple variable groupsSecurityC
    2026-082.19: workflows (Pro and Enterprise), Docker and Kubernetes executors, JWT identity tokensOrchestrationP
    2026-08Ansible collection for managing Semaphore UI as codeEcosystemC
    2026-10MCP server listed in all editions; 2.20 alphas with an audit UIAIV
    • Centre of gravity: Security (3), Scale (2), Orchestration (2)
    • Blind spots (no items): Language, Bare metal, Licensing
    • Driver mix: Vision 8%, Customer need 69%, Market window 8%, Upstream 0%, Portfolio 15%
    • Idea-to-need index: 0.09 (0 = demand-pulled, 1 = thesis-pushed)

    Credibility: 17.0 / 25

    CriterionScoreBasis
    Cadence3.5A minor release every one to three months since 2024-10, with a six-month gap before 2.17; two lines patched; no published rhythm
    Roadmap transparency4Public roadmap page with target versions, open repository and release notes; no status labels or dates
    Say-do3.5The one dated promise checked (workflows in 2.19) shipped in 2.19 on 2026-08-15; the 2.20 items are still alpha; thin record
    Velocity3.5One or two substantive features in most minors since 2.11
    Lifecycle stability2.5No support policy, Pro terms changed at least once (legacy pricing clause), and the project depends on one maintainer

    Commercial risk: 9.0 / 25 (lower is better)

    CriterionScoreBasis
    Price and licence volatility2MIT unchanged; the Pro price model changed at an unknown date (legacy pricing clause) and workflows are listed as Community on one page and Pro on another
    Purchase constraints1Free Community edition, monthly or annual Pro, 30-day trial; capacity cannot be reduced during a term
    Channel and access1Open downloads on GitHub and Docker Hub; sales direct through the site; no partner programme
    Owner stability3.5No legal entity named; one main author holds most commits
    Cost of staying supported1.5Updates are free; support SLA only on paid plans; HA and audit features sit in Enterprise

    Technical lock-in: 3.0 / 25 (lower is better)

    CriterionScoreBasis
    Formats0.5Playbooks and Terraform code stay in the operator's Git repositories
    Export path1Project export and import via CLI since 2.17 and a REST API; objects need recreating in another controller
    Stack coupling0.5SQLite, MySQL or PostgreSQL; Redis for Enterprise HA
    Hardware / cloud coupling0Any host; Docker image and Helm chart
    Skills and tooling coupling1Skills are those of the wrapped tools

    Integrator fit: 7.0 / 20

    CriterionScoreBasis
    Partner programme openness1No partner programme found; a partnership contact address only
    Multi-tenancy and self-service2.5Projects, teams, roles and survey inputs; custom roles and IdP group mapping in Enterprise; no tenant isolation described
    API and infrastructure-as-code3REST API with Swagger docs, CLI and a vendor Ansible collection
    Skills and certification0.5None found

    AI leverage: 3.0 / 20

    How the product uses AI, scored per year like capability. Only GA counts.

    Criterion20192022TodayLatest step
    AI code and plan generation0002015: None
    AI platform stack0002015: None
    AI-assisted operations0002015: None
    Agent openness2232026: Official MCP server for projects, templates, inventories, repositories, schedules and task context, listed in all editions without a GA or preview label

    Year by year: 2015: 2 · 2016: 2 · 2017: 2 · 2018: 2 · 2019: 2 · 2020: 2 · 2021: 2 · 2022: 2 · 2023: 2 · 2024: 2 · 2025: 2 · 2026: 3

    AI say-do, 2023 to 2026: 1 AI items announced; 0 GA, 1 still preview, 0 pending, 0 dropped. The MCP server is the only AI item; it is counted as preview until the vendor labels it.

    Semaphore UI has no AI generation or assistant. Its MCP server lets an agent manage projects and launch tasks, which is more than most tools in its price class offer, but the documentation could not be read and no status label is published.

    Reading

    Semaphore UI scores as what it is: a thin, fast-moving control layer whose provisioning and configuration depth belong to the tools it launches. Since late 2024 it has added the pieces a small team needs from a controller, namely runners, SSO, secret managers, workflows with approvals and executors, and its public roadmap is more specific than Red Hat's. The gaps are structural: no dry run or drift view of its own and no audit log outside an alpha. The heaviest risk is commercial, because no company stands behind the Pro licence and one maintainer carries the project. The buyer's check is simple: confirm the entity you are paying and whether the features you need, workflows and HA among them, sit in Community, Pro or Enterprise.

    Open questions and items to re-score

    • First release year and the legal entity behind Pro billing were not found; scores before 2024 are inferred.
    • MCP server status (GA or preview) and its documentation: the docs pages returned HTTP 403.
    • Workflows are listed under Community on the pricing page and under Pro on the docs, roadmap and release page.
    • Whether OIDC sign-in was free before it was listed under Pro; a 2023-07-23 vendor post describes OpenID setup.
    • The pricing page shows node tiers up to 5,000 next to a stated 500-node cap per Pro licence.
    • Dates of the 2.19.7 approval step and the 2.19 security and executor features come from the vendor's release page, not from tagged GitHub dates.

    Related reading

    More from Infrastructure Automation Ledger 2026