ZeroShot Pricing 2026

Open-source agent observability where plain-English Signals catch failures you never instrumented

Pricing: Free is $0/month with 1 GB of data and no overage, $5 in Signals with no overage, 7-day retention, 1 project, 1 seat and community support. Starter is $30/month with 3 GB included then $2/GB, $15 in Signals then $0.5 per 1M input and $3 per 1M output tokens, 30-day retention, unlimited projects and seats and email support. Pro is $150/month with 10 GB included then $1.50/GB, $50 in Signals then $0.4 per 1M input and $2.5 per 1M output tokens, 6-month retention, unlimited projects and seats and Slack support. Enterprise is custom with on-premise deployment and dedicated support. SOC 2 Type II, HIPAA and server-side PII removal are listed on the paid tiers.

A free, self-hosted dashboard that visualises Claude Code, Codex, Gemini CLI, Aider and other agents as pixel characters, with per-turn token, cost and tool-call inspection.

Pricing: Free forever, with no login required. The daemon is self-hosted and runs locally, started with `npx @bulletproof-sh/ctrl-daemon@latest`; the web app is standalone and needs no IDE. The vendor's own comparison table lists ctrl's price as free forever against a rival VS Code extension that is free with a $2 tileset. No paid tier is advertised.

Managed skills platform for AI agents — server-side prompts, tools and memory with tracing, evals and global inference

Pricing: Three tiers. Free is $0 to start on pay-as-you-go pricing, including managed skills and memory, access to cost-efficient models, geo-distributed LLM providers with low latency, 1-day API access log retention, email support and standard rate limits; approved open-source maintainers can request $200/month in credits. Growth is $25/month and includes everything in Free plus $25 in token credits each month with pay-as-you-go beyond that, higher rate limits for production workloads, access to the latest frontier models (the page names Fable 5, Claude Opus 4.8, GPT-5.5 and DeepSeek V4 Pro), unlimited managed skills and memory, Agent Insights reporting on what customers ask and what it costs, overdraft protection so agents keep running past the balance, 90-day API log retention and advanced support. Enterprise is custom-priced with SSO/SAML, custom SLAs, VPC and private deployments, audit logs, granular access controls, security reviews and volume pricing. The operator is Allegro US, LLC in San Francisco.

Agent cloud handling hosting, scaling, evals and observability for agents on any model or harness

Pricing: Individuals is $0/month with $25 in free credits, unlimited deployed agents, 7 days of data retention, and versioning, observability and community support. Teams is $500/month with $500 in credits included, unlimited deployed agents, 30 days of data retention, and adds orchestration, evaluations and priority support. Enterprise is custom-priced with custom credits, unlimited deployed agents, custom data retention, SSO, RBAC, hybrid deployment and dedicated support. Across all tiers you pay only for what you run beyond included credits. The product is generally available; agents are defined in a short YAML file specifying model and harness, both of which are intended to be swappable.

Two-line proxy that kills runaway agent loops with runtime budget enforcement and anomaly detection

Pricing: Free: 50,000 requests/month, runaway-loop detection, budget enforcement, unlimited sessions and a basic dashboard, no credit card. Starter is $39/month for 150,000 requests/month, adding advanced analytics, edge prompt caching, custom anomaly rules, team budgets and Slack webhook alerts. Pro is $99/month for 500,000 requests/month, adding agent-level budgets, audit-log retention, SLA monitoring and shadow-AI discovery. Enterprise is custom, with unlimited requests, a 99.9% SLA, SOC 2 compliance exports, custom data residency, SSO via SAML/Okta, Azure OpenAI and AWS Bedrock support and a dedicated account manager. Model usage itself still bills to your own provider account.