PAGES
← all tools
◆ AI Runtime Defense & Guardrails

Certiv

Token Spend Is a Potential Signal for a Rogue Agent commercial

Certiv is an AI agent assurance platform for endpoints that discovers every agent, monitors activity in real time, and enforces policy before risky actions execute.

source ↗ rss ↗ What is a cyber toolchain?

Summary

Certiv is a commercial AI security product that enforces spend and access controls on AI agents running across company endpoints, deployed as a single install called Scout that works alongside existing device management tools. Beyond discovery, it lets teams set per-endpoint dollar budgets against negotiated contract rates, with Report, Notify, and Block modes that let a security or IT team watch consumption before deciding to stop a runaway agent's prompts at the endpoint; a simulation feature checks a proposed threshold against historical usage first, so enforcement doesn't surprise anyone. A separate Privacy Mode keeps compliant agent sessions hidden from security teams by default, logging who accesses one and why, which points to compliance and governance owners as much as security staff.

Certiv is an AI agent assurance platform for endpoints that discovers every agent, monitors activity in real time, and enforces policy before risky actions execute.

How Certiv helps

Which AI agents are running on our endpoints right now, and what are they actually doing?

a single endpoint install builds a live catalog of agents, tools, MCP servers, and LLMs in use, scored for risk and updated as sessions run, including agents nobody registered or approved

Is one of our agents currently being manipulated or exfiltrating data?

the policy engine reads the session as a whole rather than isolated events, flagging deviations like out-of-scope credential calls or data staged for an external post, and can block the action before it executes

Which agent sessions actually need a security reviewer's attention today?

with Privacy Mode enabled, only sessions that tripped a policy violation are surfaced automatically, while compliant sessions stay hidden until someone records a justified reason to view them

Who looked at a private, compliant agent session, and why?

every access to a session hidden by Privacy Mode is logged with the reviewer's identity and their recorded justification, giving compliance and works councils an audit trail

Which endpoints or agents are burning through model spend faster than expected?

session-level token attribution breaks down cost by agent, endpoint, user, and model with historical trend views, so an abnormal consumption pattern is visible before it becomes a budget problem

Which endpoint would have blown its budget if we'd enforced a limit last month?

budget simulation replays historical usage against a proposed per-endpoint dollar threshold, showing how often it would have fired before you switch that endpoint into blocking mode

all 6 features, with the evidence for each →

Languages

Not documented

Features

6 capabilities

Built from Certiv's releases, docs, README and our own runs. “Seen” marks the month we first spotted a release the vendor didn't date. Group names and tags are our labels.

Compare with peers →

What Certiv offers you

All capabilities 6 capabilities
Token Budgeting shipped Enforces per-endpoint dollar budgets with configurable per-model rates and three enforcement modes: Report, Notify, or Block. 9 releases · 3 other sources · first seen Sep 2026

release 9 total

  • Adds ability to set dollar budgets per endpoint with configurable enforcement actions: Report, Notify, or Block. Token Spend Is a Potential Signal for a Rogue Agent · Sep 2026 · source · release history
  • Adds support for entering contract-accurate rates per model in Certiv Cost instead of relying on list pricing. Token Spend Is a Potential Signal for a Rogue Agent · Sep 2026 · source · release history
  • Supports custom contract rates entered per model so budget thresholds reflect what your organization actually pays rather than list pricing. Token Spend Is a Potential Signal for a Rogue Agent · Sep 2026 · source · release history
  • Supports configurable custom rates per model so budget thresholds reflect negotiated pricing rather than list pricing. announced · Sep 2026 · source · release history
  • Adds support for entering negotiated per-model rates so budget thresholds are computed against actual contracted pricing rather than list pricing. announced · Sep 2026 · source · release history
  • Adds per-endpoint dollar budgets with three enforcement modes: Report (logs overage, no interruption), Notify (fires alerts, work continues), and Block (stops future prompts at the endpoint before the request leaves). Token Spend Is a Potential Signal for a Rogue Agent · Sep 2026 · source · release history

example

  • Switch to Block mode on high-risk endpoints once you have a baseline, so a runaway agent session is stopped at the endpoint before the prompt leaves and drains shared quota. In the Certiv console, go to Cost > Endpoints, select the endpoint, and change the enforcement mode from 'Notify' to 'Block' — future prompts exceeding the budget are stopped before the request is… Token Spend Is a Potential Signal for a Rogue Agent · Sep 2026 · source
  • Start monitoring token spend without interrupting work — switch to Block only after you understand normal consumption patterns for an endpoint. In the Certiv console, go to the endpoint you want to protect, open Cost settings, set a dollar budget for your chosen period, and set the enforcement mode to 'Notify'. Token Spend Is a Potential Signal for a Rogue Agent · Sep 2026 · source
  • Contain a potentially runaway agent by enforcing a hard spending ceiling on a single endpoint — stopping future prompts before they leave. In Certiv Cost, select the target endpoint, set a dollar budget for your chosen period, and set the enforcement mode to Block. announced · Sep 2026 · source
Privacy Mode shipped Controls session visibility by defaulting compliant sessions to private and surfacing only policy-violating sessions, while logging justified access for audit trails. 8 releases · 1 other source · first seen Jul 2026

Also counts toward: Administrative audit log

release 8 total

  • New Privacy Mode changes session visibility defaults so only sessions with policy violations are automatically surfaced to security teams, while compliant sessions stay private unless an authorized user records an explicit justification for access. Certiv screenshot Enterprise Security with Privacy by Design: Introducing Privacy Mode · seen Aug 2026 · source · release history
  • Privacy Mode integrates with the existing Policy Engine and Scout endpoint observer: policy verdict determines session visibility without altering real-time Block, Pause, and Steer interventions. Enterprise Security with Privacy by Design: Introducing Privacy Mode · seen Aug 2026 · source · release history
  • Certiv logs who accessed a private compliant session and the recorded reason, creating an auditable trail for governance, privacy teams, and works councils. Enterprise Security with Privacy by Design: Introducing Privacy Mode · seen Aug 2026 · source · release history
  • Privacy Mode integrates with the existing Policy Engine and Scout endpoint observer — Block, Pause, and Steer runtime interventions continue operating unchanged; Privacy Mode governs only post-session review access. Enterprise Security with Privacy by Design: Introducing Privacy Mode · Jul 2026 · source · release history
  • Adds Privacy Mode feature that keeps compliant AI agent sessions private by default while surfacing violations automatically to security teams Enterprise Security with Privacy by Design: Introducing Privacy Mode · Jul 2026 · source · release history
  • Introduces Privacy Mode (now in Preview), which restricts session visibility by default so that only agent sessions with policy violations are automatically surfaced to security teams. Enterprise Security with Privacy by Design: Introducing Privacy Mode · Jul 2026 · source · release history

example

  • Enable Privacy Mode in a proof of value to observe AI agents without broad surveillance concerns. Contact your Certiv team to enable Privacy Mode for the Proof of Value Enterprise Security with Privacy by Design: Introducing Privacy Mode · Jul 2026 · source
Budget Simulation shipped Uses historical usage data to project how often endpoints would hit proposed spending thresholds before enforcement. 4 releases · 2 other sources · first seen Sep 2026

release

  • Adds a budget simulation feature that uses historical usage to project how often an endpoint would have hit a proposed threshold before you enforce it. Token Spend Is a Potential Signal for a Rogue Agent · Sep 2026 · source · release history
  • Adds a budget simulation feature that shows how often an endpoint would have hit a proposed threshold using historical usage, making it easier to calibrate limits before enforcing them. announced · Sep 2026 · source · release history
  • Adds budget simulation using historical usage data, letting teams preview how often a proposed threshold would have been hit before switching to enforcement. announced · Sep 2026 · source · release history
  • Adds budget simulation using historical usage to preview how often an endpoint would have hit a proposed threshold before enforcing it. Token Spend Is a Potential Signal for a Rogue Agent · Sep 2026 · source · release history

example

  • Validate a proposed budget ceiling against real historical traffic before enforcing it, to avoid blocking legitimate work. In Certiv Cost, open the budget simulation view for the endpoint, enter a proposed dollar threshold, and review how many times that ceiling would have been triggered in the historical usage period. announced · Sep 2026 · source
  • Use the budget simulation against historical usage to pick a threshold that would have caught past overruns without generating noise before switching an endpoint to Block mode. In the Certiv console, go to Cost > Endpoints, select an endpoint, enter a proposed dollar budget, and run the simulation — Certiv Cost shows how many times that ceiling would have been triggered… Token Spend Is a Potential Signal for a Rogue Agent · Sep 2026 · source
End-User Spend Visibility shipped Allows end-users to view their own per-endpoint token consumption broken out by hour via the Scout interface. 4 releases · first seen Sep 2026

release

  • Adds a Scout view giving end users visibility into their own per-endpoint token consumption broken out by hour, without requiring administrator access. announced · Sep 2026 · source · release history
  • Adds end-user spend visibility in Scout, showing each user their own token consumption on their own endpoint broken out by hour. Token Spend Is a Potential Signal for a Rogue Agent · Sep 2026 · source · release history
  • Adds user-level spend visibility in Scout Cost, showing individual consumption broken out by hour on their own endpoints. Token Spend Is a Potential Signal for a Rogue Agent · Sep 2026 · source · release history
  • Adds a Scout end-user spend view showing each user their own token consumption on their own endpoint, broken out by hour. Token Spend Is a Potential Signal for a Rogue Agent · Sep 2026 · source · release history
Agent Cost Attribution shipped Provides session-level token tracking by agent, endpoint, user, and model with historical trend views to identify abnormal consumption. 3 releases · first seen Sep 2026

release

  • Adds Certiv Cost in public preview for token usage visibility across all agents, breaking down spend by model, endpoint, user, and session. Token Spend Is a Potential Signal for a Rogue Agent · Sep 2026 · source · release history
  • Adds Certiv Cost (public preview) with session-level token attribution — tracking spend by agent, endpoint, user, and model — plus historical trend views to surface abnormal consumption. Token Spend Is a Potential Signal for a Rogue Agent · Sep 2026 · source · release history
  • Adds session-level token attribution tracking spend by agent session, endpoint, user, and model with historical trend data. Token Spend Is a Potential Signal for a Rogue Agent · Sep 2026 · source · release history
AI Usage Discovery documented Detects and surfaces unsanctioned AI agents, models, and tools to enable governance before unauthorized usage creates risk. 1 other source

docs

  • Any Agent, Model, or Tool Minutes To Deploy Discover and govern unsanctioned AI before it becomes a security event seen Oct 2026 · source

Lines in monospace are the tool's own words: help text parsed from its source, or an endpoint from its API document. Under release, a line is our summary of a dated release. Under docs, it is quoted from the tool's own documentation or website. Each links back to the source it came from.

Command line

1 command for Certiv, each with where it came from. This tool has no source analysis yet, so there is no option reference and no captured run beside them.
Source

Screenshots

From the release notes, newest first. Click one to see it full size.

Enterprise Security with Privacy by Design: Introducing Privacy Mode seen Aug 16, 2026

Score card

  1. Shipping velocity releases a month, last 12 months
  2. Documentation quality release-notes rubric, 0–100
  3. Breaking-change discipline how rarely a release breaks users, and notice given
  4. Maintenance risk quiet time against its own cadence, maintainers
  5. Innovation & surface new capabilities, API and MCP surface

Five scores for Certiv, each built from the releases, notes and API history this site tracks — and, on Researcher and Business, where it stands among every measured tool in its category and how that has moved.

Score cards are part of every paid plan.

See plans →

Each score is 0–100 on a published rubric, derived from data this site already shows. A dimension we cannot measure says so — it is never scored 0 and never enters a rank.

How this is scored
  • Shipping velocity — releases a month over the last 12 months, on a log scale where 16 a month scores 100. Needs 3 releases in the window.
  • Documentation quality — the release-notes rubric published on the analytics page, unchanged.
  • Breaking-change discipline — the share of releases carrying a breaking change (none scores 100, half or more scores 0). Only projects that label breaking changes are measured: a clean record cannot be told from an unlabelled one.
  • Maintenance risk — quiet time measured against the tool's own release cadence (quiet longer than 90 days and three typical gaps scores 0), plus the number of active maintainers and whether contributors are growing, where known. Higher is lower risk.
  • Innovation & surface — new capabilities in the feature ledger this year, API and MCP operations added, and the share of commits that are features: the mean of whichever we hold.
  • Community health (open-source tools only) — each is a direction, not a size: contributors this year against last, the bus factor now against a year ago, and issues closed per issue opened, where steady scores 50, doubled 100 and halved 0; plus the median time to a maintainer's first reply (an hour scores about 100, a month or more 0). A commercial card has no repository and carries none of the four.
  • Documented delivery (commercial tools) — when a vendor's documented new capabilities have slowed against its own baseline, that counts against the overall standing as heavily as two low dimensions. Steady delivery, or too little history to judge, leaves the standing alone; it is never a sixth score.
  • Open-source tools are scored on repository signals; commercial tools on their published changelog. Ranks compare tools within one category and one profile and need at least five measured peers; the overall standing needs three ranked dimensions. Best-in-breed or consolidator is a shape, not a quality, and is never scored.

Release-note detail over time

See how much detail Certiv puts in its release notes, quarter by quarter, and compare it with tracked tools in the same category. This history uses up to ten retained, reviewed notes per tool; sparse or blank older quarters may reflect that limit.

Checking release-note history…

Explore the full chart →

Website history

certiv.ai, every change

Certiv's home page and 16 more pages as they looked each month from April 2026 to October 2026: 19 renders, 1 change of tagline or palette. Step through them, play them, or put any two months side by side.

Website history is included with Researcher and Business.

See plans →

Releases

Show

5 entries

  1. September 20262
  4. August 20262
  7. July 20261
    Show

    5 entries

    1. announced Sep 4, 2026 · issue 070

      Certiv: Introduces Certiv Cost (now in public preview), providing session-level token attribution broken out by agent, endpoint, user, and model with historical trend data.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

      • Adds per-endpoint dollar budgets with three enforcement modes — Report, Notify, and Block — where Block stops prompts at the endpoint before they are sent.
      • Supports configurable custom rates per model so budget thresholds reflect negotiated pricing rather than list pricing.
      • Adds a budget simulation feature that shows how often an endpoint would have hit a proposed threshold using historical usage, making it easier to calibrate limits before enforcing them.
      • Adds a Scout view giving end users visibility into their own per-endpoint token consumption broken out by hour, without requiring administrator access.

      4 details in Everything

    2. Token Spend Is a Potential Signal for a Rogue Agent Sep 4, 2026 · issue 068

      Certiv Cost enters public preview with session-level token attribution, per-endpoint budgets, and end-user spend visibility in Scout.

      └──▷ HOW TO FIND IT
      Start in Notify mode to understand real consumption patterns across your endpoints before committing to hard limits that could block legitimate work.
      📍In the Certiv console, go to Cost › Endpoints, select an endpoint, set a dollar budget for your desired period, and choose Notify as the enforcement mode — alerts will fire when the threshold is crossed but prompts will continue to go out.
      Switch to Block mode on high-risk endpoints once you have a baseline, so a runaway agent session is stopped at the endpoint before the prompt leaves and drains shared quota.
      📍In the Certiv console, go to Cost › Endpoints, select the endpoint, and change the enforcement mode from Notify to Block — future prompts exceeding the budget are stopped before the request is sent to the model.
      Use the budget simulation against historical usage to pick a threshold that would have caught past overruns without generating noise before switching an endpoint to Block mode.
      📍In the Certiv console, go to Cost › Endpoints, select an endpoint, enter a proposed dollar budget, and run the simulation — Certiv Cost shows how many times that ceiling would have been triggered over the historical period.
      • Adds Certiv Cost (public preview) with session-level token attribution — tracking spend by agent, endpoint, user, and model — plus historical trend views to surface abnormal consumption.
      • Adds per-endpoint dollar budgets with three enforcement modes: Report (logs overage, no interruption), Notify (fires alerts, work continues), and Block (stops future prompts at the endpoint before the request leaves).
      • Supports entry of negotiated per-model rates so budget thresholds are computed against what your organization actually pays rather than list pricing.
      • Adds a budget simulation feature that uses historical usage to project how often an endpoint would have hit a proposed threshold before you enforce it.
      • Adds end-user spend visibility in Scout, showing each user their own token consumption on their own endpoint broken out by hour.

      5 details in Everything

    3. 1.0.0 Aug 16, 2026 · issue 047

      Certiv now publishes an API — 2 endpoints across 1 area: Lead Capture

      • Lead Capture (2 endpoints) — Public endpoints for requesting a demo or subscribing to the newsletter.

      1 detail in Everything

    4. Enterprise Security with Privacy by Design: Introducing Privacy Mode seen Aug 16, 2026 · issue 047

      Certiv introduces Privacy Mode, making agent session visibility proportional to risk with auditable access for compliant sessions.

      • New Privacy Mode changes session visibility defaults so only sessions with policy violations are automatically surfaced to security teams, while compliant sessions stay private unless an authorized user records an explicit justification for access.
      • Certiv logs who accessed a private compliant session and the recorded reason, creating an auditable trail for governance, privacy teams, and works councils.
      • Privacy Mode integrates with the existing Policy Engine and Scout endpoint observer: policy verdict determines session visibility without altering real-time Block, Pause, and Steer interventions.
      • Privacy Mode is now available in Preview to existing Certiv customers — ask your Certiv team to enable it.

      4 details in Everything

    5. Enterprise Security with Privacy by Design: Introducing Privacy Mode Jul 27, 2026 · release notes

      Certiv Enterprise introduces Privacy Mode for AI agent monitoring, making session visibility proportional to risk and keeping compliant sessions private by default.

      └──▷ HOW TO FIND IT
      Enable Privacy Mode in a proof of value to observe AI agents without broad surveillance concerns.
      📍Contact your Certiv team to enable Privacy Mode for the Proof of Value
      • Adds Privacy Mode feature that keeps compliant AI agent sessions private by default while surfacing violations automatically to security teams
      • Enables audited, justification-required access for governance team members to view otherwise private sessions

      2 details in Everything

    Add the badge to your README

    Maintain this project? The badge links readers straight to this page, so they can follow every release here.

    Tracked by Cyber Toolchain Choose a badge style and copy the code →
    compare.md

      Free the public facts, no sign-in Any paid plan all other sections Researcher & Business pricing history and category rank

      my-toolchain — 0 tools

        Betas, release candidates and previews. Turn it off to see final releases only. It also applies to changes you ask for over MCP when you're signed in.
        browse all tools → score card for your toolchain →
        cyber-toolchain / feedback

        Send us feedback