# Pickaxe — No-code platform for building AI agents and selling them through branded portals with billing built in. > Source: The Agents Index — https://theagentsindex.com/pickaxe (structured, researched, re-verified) > Facts last verified: 2026-09-09 Pickaxe is a no-code platform for building AI chat and form agents on hosted frontier models, grounded in your documents and wired to third-party apps through Actions. Builders deploy them to portals, embeds, messaging channels or an API and charge end users for access. | Fact | Value | | --- | --- | | Website | https://pickaxe.co | | Price | from $37/month billed monthly, $29/month billed annually | | Free tier | Yes | Free account with $5 of usage that replenishes every 30 days from signup; paid credits and plan features (extra workspaces, unlimited actions, included API) require Gold at $37/month or above. ## What it does **Done-for-you service** · Partly — Self-serve software first, but the vendor's own team will scope and build your workspace in paid one-to-one sessions starting at $30 after a free 15-minute consultation. > Custom Support is hands-on, one-on-one help from the Pickaxe team. > — https://pickaxe.co/learn/custom-support · Custom Support docs page, 'What is Custom Support?' section ## How much it does unattended **Runs autonomously** · Partly — Standard agents only act when a user writes to them; unattended scheduled runs arrive with Agentic Runtimes, which the docs mark as beta and gated behind a call with the vendor. > agents can run on a schedule, kicking off tasks automatically at set intervals (hourly, daily, weekly, or any cron expression) without anyone needing to start the conversation > — https://pickaxe.co/learn/openclaw-ai-engine · Agentic Runtimes docs page, 'Built-in capabilities' list, scheduled and recurring tasks bullet **Multi-agent** · Yes — Agents can be chained: one agent is connected as another's Action so a triage agent routes to specialists, which is also the vendor's recommended way past its four-action guidance. > create a waterfall setup — a primary agent that routes to other agents, each with their own focused set of actions > — https://pickaxe.co/learn/actions · Actions docs page, 'How many actions should an agent have?' section **Agent permissions** · Partly — Control is up-front rather than interactive: you scope what an agent may do with trigger prompts and by choosing whether an Action runs through your account or the end user's, with no documented approval prompt before it fires. > Each action has its own trigger prompt where you specify the exact conditions that should cause it to fire. > — https://pickaxe.co/learn/actions · Actions docs page, 'How actions work' section, Trigger prompt bullet ## Which models it runs on **Claude** · Yes — Anthropic is one of the providers in the model catalogue and Claude models appear as selectable options per agent, billed through Pickaxe credits rather than your own Anthropic account. > Build with OpenAI, Google, Anthropic, xAI, DeepSeek, Moonshot AI, and more, all in one platform. > — https://pickaxe.co/models · Models page, hero copy above the model grid **GPT** · Yes — OpenAI is named first among the catalogue's providers and GPT and ChatGPT model cards are selectable in the builder, with cost per model shown on the same page. > Build with OpenAI, Google, Anthropic, xAI, DeepSeek, Moonshot AI, and more, all in one platform. > — https://pickaxe.co/models · Models page, hero copy above the model grid **Open models** · Yes — Open-weight families such as GLM, DeepSeek and Kimi are offered in the catalogue, but they run on Pickaxe's infrastructure and are billed in credits, so you are not running the weights yourself. > Z.ai's flagship open model for complex reasoning, coding, tool use, and long-context agent workflows. > — https://pickaxe.co/models · Models page, model grid, GLM 5.3 card description **Model choice** · Yes — You choose the model per agent and can swap it at any time without rebuilding the agent, and the vendor makes not being locked to one provider an explicit selling point. > Building on Pickaxe means you're never locked in — when a better or cheaper model comes along, you can switch to it in seconds without changing anything else about your agent. > — https://pickaxe.co/learn/models · Models docs page, 'Why model flexibility matters' section ## Where you use it **In your editor** · Partly — There is no Pickaxe editor extension; instead the hosted MCP server exposes the workspace as tools inside Claude Code, Codex, Cursor or VS Code, which needs a workspace API key that Gold bills per request. > Connect Claude Code, Codex, Cursor, VS Code, or any other MCP client straight to your Pickaxe workspace. > — https://pickaxe.co/learn/mcp-server · Pickaxe MCP Server docs page, opening paragraph **On the command line** · Yes — A published CLI drives the whole account from a terminal, creating agents, syncing knowledge bases and configuring deployments, and is built for a coding agent to run on your behalf. > A powerful command-line interface that gives you full control over your Pickaxe account. > — https://pickaxe.co/cli · Pickaxe CLI page, 'The Pickaxe CLI' section **In a browser** · Yes — Everything is built in a browser: a visual agent builder plus hosted portals, pages and direct links your customers open on the web, optionally on your own domain. > Design, configure, and deploy custom AI agents in minutes. Set role instructions, choose from 50+ models, connect knowledge bases, and add actions — all from an intuitive visual editor. > — https://pickaxe.co/features · Platform Features grid, 'No-Code AI Agent Builder' card ## What it costs to run **How it meters** · Yes — Two meters at once: a monthly plan fee that includes a dollar-denominated credit allowance, then consumption billed per agent response at live per-model rates. > $1 credit equals $1 of usage cost. > — https://pickaxe.co/pricing · Pricing FAQ, 'What is a credit?' answer **Free tier** · Yes — You can create an account and build without paying or talking to sales; the free account carries a small monthly usage allowance rather than a countdown trial. > Every user gets $5 of usage when they sign up for a free account. This allowance replenishes monthly, on the 30-day anniversary of their sign-up date. > — https://pickaxe.co/post/how-do-credits-on-pickaxe-work · Credits explainer linked from the pricing FAQ, 'How much usage do I get?' section **API access** · Partly — A REST endpoint per agent exists on every paid plan, but on entry-level Gold each call is billed at a penny on top of the plan, and it is only included from Pro upward. > Agent & Workspace API at 1¢/req > — https://pickaxe.co/pricing · Plan cards, Gold column feature list (Pro card reads 'Included Agent & Workspace API access') **MCP server** · Yes — Another agent can drive the workspace through a hosted MCP server at a single address, authenticating with a workspace, personal or deployment key, and can run any agent without deploying it first. > Connect Claude Code, Codex, Cursor, VS Code, or any other MCP client straight to your Pickaxe workspace. > — https://pickaxe.co/learn/mcp-server · Pickaxe MCP Server docs page, opening paragraph **Bring your own key** · No — Model inference is always resold through Pickaxe credits: the vendor rules out bringing your own model keys, so you cannot put spend on your own provider account. > Pricing flows through Pickaxe credits — no BYOK, no surprise model-provider invoices. > — https://pickaxe.co/ai-agent-api · Agent API page, 'What is the Pickaxe Agent API?' FAQ answer ## Buying it for a team **A company can buy it** · Yes — A Business tier is sold to companies with collaborators, custom credit packages and dedicated support, though it is bought through a call and carries a six-month minimum commitment. > For companies and teams building custom solutions together. > — https://pickaxe.co/pricing · Plan cards, Business column subtitle **Seat model** · Partly — Not seat-priced: a flat plan fee with a collaborator ceiling per workspace, one on Gold, four on Pro, five or more on Business, so extra builders mean a tier change rather than an extra seat. > Include up to 1 collaborator per Workspace > — https://pickaxe.co/pricing · Plan cards, Gold column feature list; Collaborators row of the 'Compare all features' table **Pooled budget** · Yes — One balance for the whole workspace: every end-user interaction is drawn from the owner's credits, which is why per-user wallets and access-group limits exist to cap them. > Yes, when an end-user of your tool uses your agent, it counts against your credits. > — https://pickaxe.co/pricing · Pricing FAQ, 'When people use my agent, are they using my credits?' answer **Admin controls** · Yes — Owners restrict what agents may do for whom: access groups decide which deployments a user reaches, each user gets a usage budget, and users can be banned by email or IP. > Assign a per-user wallet — a usage budget in credits or uses. > — https://pickaxe.co/learn/member-access-groups · Member Access Groups docs page, 'What you can do with member access groups' list **Audit log** · Partly — What is recorded is end-user activity: conversations, action runs and spend per user, which the vendor calls an audit trail. No immutable log of administrative actions is documented. > Complete interaction tracking, usage pattern monitoring, and detailed analytics. Maintain the audit trails your compliance team needs for procurement reviews. > — https://pickaxe.co/enterprise-ai-security · Enterprise AI Security page, 'Full visibility into AI activity' card **Single sign-on** · Partly — SSO is tier-gated and end-user-facing: signed-token Embed SSO from Pro, 'advanced SSO' only on Business, and the vendor never names SAML or OIDC on any page. > Advanced SSO & user management > — https://pickaxe.co/pricing · Plan cards, Business column feature list (Embed SSO row of the 'Compare all features' table shows Pro and Business only) ## What happens to your code **Opt out of training** · Partly — The terms reserve the right to train on prompts and outputs and exempt only Pro members for as long as they stay on Pro; uploaded documents are always excluded, and there is no self-serve toggle. > Pickaxe will not train models on any data generated by members of the Pro tier during the time that they retain Pro tier membership. > — https://pickaxe.co/terms · Terms and Conditions, section 17 'Model Training' **Getting out** · Partly — The CLI pulls a local copy of agents, deployments and documents, and bought credits never expire, but plan credits reset each cycle without stacking and nothing is refundable on cancellation. > Any additional credits that you buy will never expire and roll over to any month. > — https://pickaxe.co/post/how-do-credits-on-pickaxe-work · Credits explainer linked from the pricing FAQ, 'Can I buy more credits?' section **Certifications** · Yes — The vendor claims a SOC 2 examination plus GDPR and CCPA compliance, with reports offered to enterprise customers through a Trust Center whose link did not resolve when tested. > SOC 2 examined, GDPR and CCPA compliant, with enterprise SSO and end-to-end encryption. > — https://pickaxe.co/enterprise-ai-security · Enterprise AI Security page, hero copy under 'Secure AI development and governance' _Not established: Data residency_ ## Provenance Every fact above comes from the vendor. Nothing here is independently corroborated yet. ## Coverage elsewhere None recorded. Every fact above comes from the vendor and is not independently corroborated yet. ## What changed - 2026-09-10 — Criteria prose - 2026-09-09 — Capabilities · Commercial terms · Free tier · Price ## Correcting this record Anything marked "Not established" is a question we have not answered — it is not a No. If you represent Pickaxe, or anything here is wrong or out of date, you can supply it with a source: the operations this site accepts from an agent are listed at https://theagentsindex.com/.well-known/agents.json — that catalogue needs no key to read, and each entry states its own auth. Supplying evidence is free. Nothing that can be bought here changes what a field says, whether a gap is closed, or how this record is scored.