A CLI tool for local-first task management and session logging.

Install

mkdir -p .claude/skills/td && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/13434" && unzip -o skill.zip -d .claude/skills/td && rm skill.zip

Installs to .claude/skills/td

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

Manage tasks and issues with the td CLI. Use for creating, tracking, querying, and handing off work items. Covers task lifecycle, session management, structured handoffs, epics, dependencies, and boards.
203 chars✓ has a “when” trigger
Intermediate

Key capabilities

  • Start new task management sessions
  • Pick and start work on issues
  • Log progress, blockers, decisions, attempts, and results
  • Create structured handoffs before ending a session
  • Submit work for review and manage approvals/rejections
  • Query issues using quick filters or TDQ query language

How it works

The td CLI manages tasks and issues locally, tracking their lifecycle, session progress, and structured handoffs. It uses commands to start sessions, log activities, and transition issues through different states.

Inputs & outputs

You give it
CLI commands with issue IDs, descriptions, log messages, or query strings
You get back
Session context, issue lists, detailed issue information, log entries, or handoff summaries

When to use td

  • Tracking dev work sessions
  • Logging task decisions
  • Managing issue dependencies
  • Structured session handoffs

About this skill

td — Task & Session Management

td is a local-first task management CLI optimized for AI-assisted development workflows. It tracks issues, sessions, structured handoffs, and progress logs.

Session Workflow

Every session follows this pattern:

  1. Start — Load context and begin a new session:

    td usage --new-session    # Full context dump (issues, logs, handoffs)
    td status                 # Quick dashboard: focus, reviews, blocked, ready
    
  2. Pick work — Find what to work on:

    td next                   # Highest-priority open issue
    td ready                  # All open issues sorted by priority
    td list --mine            # Issues assigned to current session
    
  3. Start an issue — Set focus and mark in-progress:

    td start <issue-id>       # Sets focus + status to in_progress
    
  4. Log progress — Track decisions, blockers, attempts:

    td log "Implemented auth middleware"              # Progress (default)
    td log --blocker "Waiting on API key from team"   # Blocker
    td log --decision "Using JWT over session tokens"  # Decision
    td log --tried "Attempted Redis caching"           # Attempted approach
    td log --result "Redis too complex, using in-memory" # Result
    td log --hypothesis "Rate limiting might fix 429s"   # Hypothesis
    
  5. Handoff — Capture structured state before ending session:

    td handoff <issue-id> \
      --done "Implemented login endpoint" \
      --done "Added JWT validation" \
      --remaining "Add refresh token flow" \
      --remaining "Write tests for auth middleware" \
      --decision "Using RS256 for JWT signing" \
      --uncertain "Should we support OAuth2 providers?"
    
  6. Submit for review — When work is complete:

    td review <issue-id>      # Submit for review (preferred)
    
  7. Review — Different session approves or rejects:

    td reviewable             # What can I review?
    td approve <issue-id>     # Ship it → closed
    td reject <issue-id> --reason "Need tests"  # Back to in_progress
    

Creating Issues

# Basic
td create "Fix login redirect bug"

# Full options
td create "Add OAuth2 support" \
  --type feature \
  --priority P1 \
  --description "Support Google and GitHub OAuth2 providers" \
  --labels "auth,backend" \
  --parent <epic-id>

# Types: task (default), bug, feature, epic, chore
# Priorities: P0 (critical) → P4 (minimal)
# Points: Fibonacci — 1, 2, 3, 5, 8, 13, 21

Querying Issues

Quick filters

td list                           # All open issues
td list --all                     # Include closed
td list --type bug                # Only bugs
td list --priority P0             # Only critical
td list --status in_progress      # In progress
td list --labels auth             # By label
td list --epic <epic-id>          # All tasks in an epic
td list --sort priority            # Sort by field
td list --long                    # Detailed output
td list --json                    # JSON output

TDQ query language

td query "status = open AND type = bug"
td query "priority <= P1"                    # P0 and P1
td query "created >= -7d"                    # Last 7 days
td query "title ~ auth OR description ~ auth"
td query "log.type = blocker"                # Issues with blockers
td query "implementer = @me AND is(in_progress)"
td query "descendant_of(<epic-id>)"          # All children of epic
td query "rework()"                          # Rejected issues

Saved boards

td board create "Active Bugs" "type = bug AND is(open)"
td board create "My Work" "implementer = @me AND NOT is(closed)"
td board list                                # Show all boards
td board show "Active Bugs"                  # Run saved query

Issue Lifecycle

open → in_progress → in_review → closed
  ↕        ↕            ↕
blocked  blocked      rejected → in_progress (rework)

Review workflow (preferred for non-trivial work)

td start <id>           # open → in_progress (sets focus)
# ... do work, log progress ...
td handoff <id> --done "..." --remaining "..."   # capture state
td review <id>          # in_progress → in_review (requires handoff)

# Different session reviews:
td reviewable           # what can I review?
td approve <id>         # in_review → closed (ship it)
td reject <id> --reason "Missing error handling"  # in_review → in_progress (back to work)

Other transitions

td block <id>           # → blocked
td unblock <id>         # → open
td close <id>           # → closed (skip review, for admin closes only)
td reopen <id>          # → open

Guidelines

  • Always use reviewapprove/reject for non-trivial work — don't skip to close
  • close is for admin purposes only: duplicates, won't-fix, trivial/minor tasks
  • Session isolation: the session that implements can't approve its own work
  • reject sends work back to in_progress for rework — always include a --reason

Dependencies

td dep add <id> <depends-on-id>     # A depends on B
td dep rm <id> <depends-on-id>      # Remove dependency
td depends-on <id>                  # What does this issue need?
td blocked-by <id>                  # What's waiting on this issue?
td critical-path <id>               # Longest dependency chain

Epics & Hierarchy

# Create an epic
td create "Auth System" --type epic --priority P1

# Add child tasks
td create "Login endpoint" --parent <epic-id>
td create "JWT validation" --parent <epic-id>

# View hierarchy
td tree <epic-id>                   # Visual tree
td list --epic <epic-id>            # All descendants
td epic list                        # List all epics with progress

Session Management

td session "working on auth"        # Name current session
td session --new                    # Start fresh session
td whoami                           # Show current session identity
td focus <id>                       # Set focus without starting
td unfocus                          # Clear focus
td resume <id>                      # Show context + set focus

Inspecting Issues

td show <id>                        # Full detail with logs, handoffs, comments
td show <id> --children             # Include child issues
td comments <id>                    # List comments
td comment <id> "Note about this"   # Add a comment

Useful Shortcuts

td status                # Dashboard: session, focus, reviews, blocked, ready
td next                  # Highest-priority open issue
td ready                 # Open issues sorted by priority
td blocked               # All blocked issues
td in-review             # Issues awaiting review
td reviewable            # Issues you can review
td deleted               # Soft-deleted issues
td restore <id>          # Restore deleted issue

Guidelines

  • Always start sessions with td usage --new-session or td status
  • Log as you go — decisions, blockers, and attempts are valuable for future sessions
  • Handoff before stopping — structured handoffs make resumption seamless
  • Use epics to group related work and track progress
  • Use dependencies to make blocking relationships explicit
  • Use boards to save frequently-used queries
  • P0 = drop everything, P1 = do next, P2 = default, P3 = backlog, P4 = someday

When not to use it

  • When a graphical user interface for task management is preferred
  • When the task management system is not local-first or CLI-based
  • When the user needs to manage tasks outside the defined lifecycle (open, in_progress, in_review, closed)

Limitations

  • Requires explicit use of `review` -> `approve`/`reject` for non-trivial work
  • The `close` command is for administrative purposes only
  • Session isolation prevents the implementer from approving their own work

How it compares

This skill provides a local-first, CLI-optimized task management system with structured session workflows and handoffs, which differs from traditional project management tools that often rely on web interfaces and less granular progress log

Compared to similar skills

td side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
td (this skill)05moReviewIntermediate
trello412moReviewBeginner
executing-plans63moNo flagsIntermediate
github-project-management46moReviewAdvanced

Try saying

Example prompts that trigger this skill in your AI assistant.

You might also like

trello

openclaw

Manage Trello boards, lists, and cards via the Trello REST API.

41205

executing-plans

obra

Use when you have a written implementation plan to execute in a separate session with review checkpoints

643

github-project-management

ruvnet

Comprehensive GitHub project management with swarm-coordinated issue tracking, project board automation, and sprint planning

427

project-clickup

incidentfox

ClickUp project management integration for incident tracking and task management

320

coo-advisor

alirezarezvani

Operations leadership for scaling companies. Process design, OKR execution, operational cadence, and scaling playbooks. Use when designing operations, setting up OKRs, building processes, scaling teams, analyzing bottlenecks, planning operational cadence, or when user mentions COO, operations, process improvement, OKRs, scaling, operational efficiency, or execution.

613

tlc-spec-driven

tech-leads-club

Project and feature planning with 4 phases - Specify, Design, Tasks, Implement+Validate. Creates atomic tasks with verification criteria and maintains persistent memory across sessions. Stack-agnostic. Use when: (1) Starting new projects (initialize vision, goals, roadmap), (2) Working with existing codebases (map stack, architecture, conventions), (3) Planning features (requirements, design, task breakdown), (4) Implementing with verification, (5) Tracking decisions/blockers across sessions, (6) Pausing/resuming work. Triggers on "initialize project", "map codebase", "specify feature", "design", "tasks", "implement", "pause work", "resume work".

77

Search skills

Search the agent skills registry