clawdoTodo list and task management for AI agents. Add, track, and complete tasks with autonomy levels β agents propose work, humans approve. Works in heartbeats, cron, and conversations. Persistent SQLite CLI with structured JSON output.
Install via ClawdBot CLI:
clawdbot install LePetitPince/clawdoYour agent has memory files, cron jobs, and chat. It has no todo list.
No way to say "do this when you get to it." Not "do this at 14:00 UTC." Not "do this right now in this conversation." Just... remember to do it. Track it. Pick it up when there's a gap.
That's clawdo.
clawhub install clawdo # installs skill + docs into your workspace
npm install -g clawdo # install the CLI binary
Requirements: Node.js β₯18
# Capture a task
clawdo add "update dependencies" --urgency soon
# Agent checks its queue (heartbeat, cron, conversation β wherever)
clawdo inbox --format json
# Agent works it
clawdo start a3f2
clawdo done a3f2 --json
add β inbox β start β done. Persistent state in SQLite. Every command has --json so agents parse structured output, not terminal art.
clawdo works everywhere agents work:
# In HEARTBEAT.md β runs every ~30 minutes
TASKS=$(clawdo inbox --format json)
AUTO=$(echo "$TASKS" | jq '.autoReady | length')
if [ "$AUTO" -gt 0 ]; then
TASK=$(clawdo next --auto --json | jq -r '.task.id')
clawdo start "$TASK" --json
# ... do the work ...
clawdo done "$TASK" --json
fi
Tasks can be tagged with permission tiers that control what the agent is allowed to do unsupervised:
| Level | Time Limit | What it means |
|-------|------------|---------------|
| auto | 10 min | Agent does it silently. Fix a typo, run tests. |
| auto-notify | 30 min | Agent does it, tells the human when done. |
| collab | Unlimited | Human required. Complex, risky, or ambiguous. |
Default: collab (safe).
Key rule: Autonomy is a permission, not a suggestion. Once set, agents can't change it. If an agent fails 3 times, autonomy demotes to collab. Safety only moves down, never up.
Agents propose, humans approve. Agent tasks always start as proposed. The human runs clawdo confirm or it doesn't happen.
# Add tasks β inline metadata parsing
clawdo add "deploy new API +backend auto-notify now"
# βββ text βββββββ βprojectβ ββlevelβββ βurgβ
# View
clawdo list # active tasks
clawdo list --status proposed # agent suggestions
clawdo next # highest priority
# Review agent proposals
clawdo confirm <id> # approve
clawdo reject <id> # reject
# Work
clawdo start <id>
clawdo done <id>
clawdo done abc,def,ghi # complete several
# Check inbox (structured)
clawdo inbox --format json
# Propose work
clawdo propose "add input validation" --level auto --json
# Execute
TASK=$(clawdo next --auto --json | jq -r '.task.id // empty')
if [ -n "$TASK" ]; then
clawdo start "$TASK" --json
# ... do the work ...
clawdo done "$TASK" --json
fi
The inbox returns: autoReady, autoNotifyReady, urgent, overdue, proposed, stale, blocked.
clawdo add "fix auth bug +backend @code auto soon"
+word β project@word β contextauto / auto-notify / collab β autonomy levelnow / soon / whenever / someday β urgencydue:YYYY-MM-DD β due datecrypto.randomInt(), no modulo biasclawdo --helpMIT
Generated Mar 1, 2026
AI agents use clawdo to manage infrastructure tasks like updating dependencies, running security scans, or deploying microservices. In heartbeat loops, they autonomously handle low-risk auto tasks, while proposing complex changes like server migrations for human approval via collab tasks, ensuring continuous delivery with oversight.
Agents integrate clawdo to queue and prioritize support requests, categorizing them by urgency and context. They autonomously resolve simple issues like password resets under auto-notify, escalating ambiguous cases to human agents as collab tasks, improving response times and reducing manual workload.
Media teams use clawdo for agents to schedule and track content creation tasks, such as drafting articles or updating social media. Agents propose editorial changes with auto levels, while major publishes require human confirmation via collab, streamlining workflows in newsrooms or marketing agencies.
In academic or scientific settings, agents manage data analysis tasks, queuing computations and report generation. They autonomously run routine analyses with auto permissions, but flag anomalies or complex interpretations as collab tasks for researcher review, enhancing efficiency in labs.
Agents utilize clawdo to monitor stock levels, reorder products, and update listings. They handle automatic restocking for low-risk items under auto-notify, while proposing pricing changes or supplier negotiations as collab tasks, optimizing inventory control with human oversight.
Offer clawdo as a cloud-based service with tiered plans based on task volume and autonomy levels. Revenue comes from monthly subscriptions, targeting small to medium businesses needing AI-assisted workflow management, with premium features like advanced analytics and team collaboration tools.
Sell customized licenses to large corporations for integrating clawdo into existing systems like CRM or ERP. Revenue is generated through one-time licensing fees and ongoing support contracts, focusing on industries like finance or healthcare that require secure, auditable task management.
Provide a free open-source version of clawdo for individual developers, monetizing through premium CLI add-ons like enhanced autonomy controls or priority support. Revenue streams include paid upgrades and consulting services for complex integrations, appealing to tech-savvy users.
π¬ Integration Tip
Start by integrating clawdo into heartbeat loops for low-risk auto tasks to build trust, then gradually expand to collab tasks for complex workflows, ensuring proper sanitization of inputs to prevent injection attacks.
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.
Search and analyze your own session logs (older/parent conversations) using jq.
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linking related objects, enforcing constraints, planning multi-step actions as graph transformations, or when skills need to share state. Trigger on "remember", "what do I know about", "link X to Y", "show dependencies", entity CRUD, or cross-skill data access.
Ultimate AI agent memory system for Cursor, Claude, ChatGPT & Copilot. WAL protocol + vector search + git-notes + cloud backup. Never lose context again. Vibe-coding ready.
Headless browser automation CLI optimized for AI agents with accessibility tree snapshots and ref-based element selection