Rakazo vs Tasklet: Local Bots or Shared Cloud Agents
Compare Rakazo and Tasklet on local execution, shared work, memory controls and the pricing behind recurring agents.
Founder, Task Machine
For a team comparing Rakazo and Tasklet, does the agent need your computer and model connection, or can it work in a managed cloud sandbox? Rakazo documents self-hosting and desktop access to your own commands and files. Tasklet's agents run in managed cloud sandboxes, and its public docs do not describe running them on your infrastructure.
For a software team, that boundary can decide the comparison before a model produces its first answer. A recurring job may depend on local files, or it may need only connected business services. Start by locating those dependencies, then compare how teammates assign work and review the result.
Persistent bots with an optional board
Rakazo gives each bot a continuous conversation, instructions, memory and a computer. Its bot guide frames the bot as a persistent participant in your work. Routines add scheduled and event-driven requests, while skills save written steps for reuse.
Its Labs board adds tickets shared by people and bots, including owners, priorities, comments and status changes. Assigning a ticket to a bot starts its work. This matters when the conversation alone is no longer enough to show who should handle a job.
The board does not establish every kind of process control. Rakazo's public docs do not describe turning a planning conversation into a reviewed execution plan before work starts. They also do not describe an enforced multi-step process with branches and conditions. Include those requirements explicitly if the job needs them.
Shared agents across threads
Tasklet organizes work around agents with instructions, connections, knowledge and automations. Its agent guide explains how teammates use shared agents in threads. Shared skills and files can give those agents reusable guidance, while agent-wide memory carries context between conversations.
This can suit a team that wants a common agent for connected business work. A thread provides a place to request a particular report or draft, and the agent can maintain context across requests. Tasklet's public docs do not describe a shared task record owned by either a person or an agent.
Sharing also defines a visibility boundary. The guide says a member of a shared agent can see and use every thread in that agent, and recommends separate agents when access needs to differ. Decide whether the agent represents an entire team or a narrower group before adding customer information.
Locate the computer and the model bill
Rakazo's desktop documentation describes running Rakazo locally or connecting bots to your computer for commands and files. Hosted computers and self-hosting are also available. Its public machine-readable overview calls Cloud beta, even though hosted plans appear on the pricing page, so confirm the current hosted-service status.
Its model connection guidance describes eligible Codex allowance sign-in and provider API keys. Rakazo recommends Anthropic API keys for Claude and warns against using Claude consumer subscription credentials for its model connection. Do not assume every consumer subscription is interchangeable with an API connection.
Tasklet's security documentation places its agents in managed cloud sandboxes. A Mac app provides access, but does not establish local agent execution. Its public pricing and MCP documentation also do not describe bringing your own AI subscription or model-provider API key.
Model choice differs within those environments. Rakazo offers a main model, per-bot and per-routine overrides, plus ordered fallbacks for failures or limits. Tasklet's credit and model guide describes Auto choosing a model for each message through Economy, Balanced or Premium settings, with specific-model choices also available. A fallback list and automatic message-level selection solve different problems.
Inspect memory and permission changes
Rakazo's memory documentation describes reading, editing, downloading or deleting bot notes, and editing shared space memory. Bots keep at most 4 KB of notes in MEMORY.md, making room by merging or removing notes. You can add more by hand or connect an optional memory service.
Tasklet describes memory spanning conversations and files, but its public docs do not describe inspecting or correcting the agent's memory. The guide also says memory is automatically enabled for new agents and cannot be turned off once enabled for an existing agent. If a correction must remain inspectable, test that requirement directly.
Rakazo's approval modes offer Ask, Auto and Never ask, editable Allow and Block lists, and approval cards in chat. Auto checks actions against your request. Broad Always allow exceptions deserve review before routines process outside text.
Tasklet sets tool permissions separately for each agent and asks before using a tool that is not enabled. You can also decide which drafts require approval. Both products therefore publish action controls, but the comparison should use the exact sending or editing tool your recurring job will invoke.
Pay for the work you intend to repeat
Rakazo's pricing page lists free self-hosting, with your own infrastructure and model connection. Hosted own-model plans start at $15 a month or $12 a month billed yearly. Rakazo credits start at $20 a month for model usage at provider prices and computer hours.
Tasklet's pricing page offers limited free trial credits and a $25 monthly plan with 10,000 credits shared across the organization. One-time credits are also available. Its credit guide says automation runs cost credits at the same rate as manual actions and recommends filtering events and reducing frequency before the AI runs.
These plans pay for different execution arrangements. Test the trigger frequency as well as the quality of a single answer, and distinguish your own model connection from vendor credits when documenting the expected bill.
Rakazo and Tasklet, point by point
The table places both vendors' answers against the same questions. Each answer includes a source and check date so you can distinguish the documented capability from a broader inference.
Both columns answer the same questions from each company's public pages. Yes means the company publishes the capability the question asks about, Partly means something narrower, No means it says it does not offer it, and Not published means its public pages do not describe it. Every answer links to its source and shows when we last checked it. The oldest check on this page was November 13, 2026.
| What you compare | Rakazo | Tasklet |
|---|---|---|
| Goals |
Rakazo
Partly
You give each persistent bot its job in chat and can schedule recurring work. Its docs do not describe an owned goal with measurable targets and progress tracking. |
Tasklet
Partly
An agent can report on revenue, support tickets and project milestones each morning. Its docs do not describe an agent owning a goal with a measurable target. |
| Leads |
Rakazo
Partly
A bot can organize other bots, rewrite their instructions and schedule their work. Its docs do not describe a project, goal or team lead responsible for reviewing progress. |
Tasklet
Partly
An agent can assign sales accounts using your rules and delegate focused work to subagents. Its docs do not describe a named lead coordinating a project, goal or team. |
| Tasks |
Rakazo
Yes
The Labs ticket board is shared by you and your bots, with owners, priorities, comments and status changes. Assigning a ticket to a bot starts its work. |
Tasklet
Partly
Agents use a task list for multi-step work, and teammates can work with shared agents in threads. Its docs describe no shared task record owned by either a person or an agent. |
| Recurring work |
Rakazo
Yes
Routines run on schedules, webhooks, GitHub events, email or Slack messages, and can watch for changes. Each routine keeps a run history. |
Tasklet
Yes
Agents run on schedules and events, including Gmail, Slack and webhooks. Its docs describe no stable root task with separate occurrence histories. |
| Workflows |
Rakazo
Partly
Skills save written steps for bots to repeat, and routines run prompts on triggers. Its docs do not describe an enforced multi-step process with branches and conditions. |
Tasklet
Partly
Shared skills describe repeatable work, and agents can request approval by email. Its docs do not establish a repeatable multi-step process that runs the same way each time. |
| Autonomy |
Rakazo
Yes
Ask, Auto and Never ask modes control actions, with editable Allow and Block lists. Approval cards collect requests in chat, and Auto checks actions against what you requested. |
Tasklet
Yes
Tool permissions are set separately for each agent, and an agent asks before using a tool that is not enabled. You can also decide which drafts require approval. |
| Learned autonomy |
Rakazo
Not published
Its docs do not describe proposing permission changes from your approval history. Always allow is a choice you make on an approval card, while Auto judges actions against your request. |
Tasklet
Not published
Tasklet lets you configure tool permissions for each agent. Its docs describe no proposals to change permissions based on approval history. |
| Memory |
Rakazo
Yes
Bots keep persistent notes and shared space memory. You can read, edit, download or delete notes, and optionally connect services for recall by meaning. |
Tasklet
Partly
Agent-wide memory carries context across threads, including conversations and files. Its docs do not describe a way to inspect or correct the agent's memory. |
| Planning |
Rakazo
Partly
Each bot has a continuous chat and an editable Open work list. Its docs do not describe turning a planning conversation into a reviewed execution plan before work starts. |
Tasklet
Partly
You can ask an agent to identify gaps and draft a shared skill for your review. Its docs do not establish a reviewed plan for a particular job before execution. |
| Intake |
Rakazo
Yes
A webhook URL and key route an outside POST request to a bot's routine. Email and Slack messages can also start routines, subject to connection and sender restrictions. |
Tasklet
Yes
An agent can receive leads through a webhook and distribute them round-robin to sales representatives. |
| Connectors |
Rakazo
Yes
The app catalog supports custom remote MCP servers, OpenAPI and GraphQL connections. Routines accept webhooks and GitHub events, while self-hosted catalog availability depends on server configuration. |
Tasklet
Yes
Tasklet connects agents to over 3,000 built-in integrations, HTTP APIs and MCP servers, and provides a sandboxed command line. Its docs describe no custom command-line connector catalog. |
| Models |
Rakazo
Partly
You choose a main model, per-bot or per-routine overrides, and ordered fallbacks for failures or limits. Its docs do not describe automatic model selection by task. |
Tasklet
Yes
Auto selects a model for each message with Economy, Balanced or Premium settings, and you can choose a specific model or an agent default. Agents can delegate parts of work to other models. |
| Cloud or your own computer |
Rakazo
Yes
Rakazo offers hosted computers and self-hosting on your server or computer. The desktop app can run Rakazo locally or connect bots to your own computer for commands and files. |
Tasklet
Partly
Tasklet agents run in managed cloud sandboxes, with a Mac app for access. Its docs do not describe running agents on your own computer or infrastructure. |
| Your own subscription |
Rakazo
Yes
You can connect provider API keys or sign in with an eligible ChatGPT plan's Codex allowance. Rakazo recommends Anthropic API keys for Claude and warns against using Claude consumer subscription credentials for its model connection. |
Tasklet
Not published
Tasklet provides models in one plan and charges Tasklet credits for agent work, including work delegated from other AI apps. Its docs do not describe using your own AI subscription or model-provider API key. |
| Pricing |
Rakazo
Yes
Self-hosting is free. Hosted own-model plans start at $15 a month, or $12 a month billed yearly. Rakazo credits start at $20 a month for model usage at provider prices and computer hours. |
Tasklet
Yes
Tasklet offers limited free trial credits and a $25 monthly plan with 10,000 credits shared across the organization. One-time credits are also available. |
| Correctness |
Rakazo
Partly
A skill taught by demonstration includes what to check, and Test runs it once. Its docs do not describe a general system that scores delivered results against defined acceptance criteria. |
Tasklet
Partly
A customer describes an agent checking live ad settings after applying approved changes, and another agent checking ads against placement and audience rules. |
For software work
| What you compare | Rakazo | Tasklet |
|---|---|---|
| Code review rules |
Rakazo
Partly
Rakazo publishes a pull-request review bot template and lets you edit bot instructions. Its docs do not establish code review against repository rules. |
Tasklet
Not published
Tasklet agents can inspect repository history and work with pull-request reviews. Its docs describe no review against repository rules. |
| Adversarial review |
Rakazo
Not published
Its docs do not describe an independent second review of risky code changes. The README says bots can check each other, and groups can use parallel helpers, without specifying that review process. |
Tasklet
Not published
Tasklet agents can work with pull-request reviews and discussions. Its docs describe no independent second review for risky changes. |
| Risk rating |
Rakazo
Not published
Its public pages do not describe rating a change's risk before merge. The GitHub catalog lists pull-request inspection and merge actions, rather than a risk assessment. |
Tasklet
Not published
Tasklet agents can inspect changed pull-request files and, with permission, merge pull requests. Its docs describe no merge risk rating. |
| Pull requests |
Rakazo
Partly
The GitHub catalog includes creating pull requests, reading reviews and replying to review comments. Its docs do not establish an agent delivery loop that revises code in response to review feedback. |
Tasklet
Yes
Tasklet can create pull requests, work with reviews and discussions, and merge with permission. Its docs describe no recorded approval of the exact commit before merging. |
A fact we cannot find or cannot check is left out, and prices can change after the date shown. Task Machine publishes this comparison and is an interested party. Rakazo and Tasklet are named only to identify them, and there is no relationship between the companies. Has something changed? Tell us.
Who each one suits
Pick Rakazo if persistent bots, editable notes and self-hosting or local computer access are central requirements. Its Labs board can give work owners, but its public docs do not describe enforced multi-step workflows or a reviewed execution plan before work starts.
Pick Tasklet if shared cloud agents, threads and automatic message-level model choice fit the team. Every shared-agent member can see its threads, and its public docs do not describe local execution or an inspectable memory correction interface.
Where Task Machine fits
Task Machine addresses the reviewed work record through Tasks owned by people or agents. You review the plan and acceptance criteria before execution, and task reviews check each requirement as Pass, Fail or Uncertain before human review. Managed Cloud and Local Workers provide a choice of agent execution location.
Task Machine is the wrong pick if you require Rakazo's free self-hosting, because its docs do not describe a free, open-source self-hosted application. See Switch from Rakazo to Task Machine. If you want shared conversational agents, compare that requirement with its Task and Workflow model. Inspect Task Machine vs Rakazo and Task Machine vs Tasklet, then inventory the files, tools and accounts needed by one recurring job to decide where its agent must run.