Task Backlog Manager

What it does

When your agent uses it

• Checking the current task backlog and what needs attention first
• Adding a new task and scoring its priority against your goals
• Reviewing and reprioritizing stale or blocked tasks
• Marking a task done and keeping the markdown export in sync

Example prompts

• “Run /tasks to see where my backlog stands.”
• “Add a task to draft the onboarding checklist and score it against my problems list.”
• “Review the backlog and tell me which tasks look stale.”

Requirements

• The sqlite3 command-line tool
• Python 3, for the db.py export script
• A workspace with tasks.db and knowledge/problems/00-overview.md
• Pre-approved tools (allowed-tools): Read, Grep, Glob, Write, Edit, Bash, AskUserQuestion

Workflow steps

8 steps, taken from the first numbered list in SKILL.md.

1. 1Parse from arguments
2. 2Get next ID: sqlite3 ~/claude-assistant/tasks.db "SELECT value FROM config WHERE key='next_id'"
3. 3Read knowledge/problems/00-overview.md to know the user's problems
4. 4Ask: which problems (by number) does this relate to? Effort (S/M/L)? Which subgoal?
5. 5Evaluate priority
6. 6On approval, insert via sqlite3
7. 7Increment next_id: sqlite3 ~/claude-assistant/tasks.db "UPDATE config SET value='T###' WHERE key='next_id'"
8. 8Re-export: python3 ~/claude-assistant/scripts/db.py export

What it can do on your machine

Read from SKILL.md and the folder at commit 546c72c. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves these tools, so the agent can use them without asking each time:

  • Read
  • Grep
  • Glob
  • Write
  • Edit
  • Bash
  • AskUserQuestion

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Shell commands in SKILL.md call:

  • sqlite3
  • python3

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Task Backlog Manager loads about 1.2k tokens when it runs. Until then it costs about 35 tokens; SKILL.md has 345 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~35

When it runs · the whole SKILL.md, loaded when a task matches

~1.2k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

• NotePre-approves every shell command (allowed-tools: Bash)SKILL.md

  allowed-tools: Read, Grep, Glob, Write, Edit, Bash, AskUserQuestion

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from mp-web3/claude-starter-kit at commit 546c72c, republished under its MIT licence (© mp-web3). 345 words, ~1,186 tokens.

name

tasks

description

Manage the task backlog — add, review, complete, and measure tasks against the 12 Favorite Problems. Run /tasks to see current status.

allowed-tools

Read, Grep, Glob, Write, Edit, Bash, AskUserQuestion

argument-hint

[status | add <description> | review | done T### | metrics]

# Active tasks sorted by priority
sqlite3 ~/claude-assistant/tasks.db -header -column \
  "SELECT id, name, priority, status, effort, problems FROM tasks WHERE status != 'done' ORDER BY CASE priority WHEN 'P1' THEN 1 WHEN 'P2' THEN 2 WHEN 'P3' THEN 3 ELSE 4 END, id"

# Blocked tasks
sqlite3 ~/claude-assistant/tasks.db -header -column \
  "SELECT id, name, blocked_by, blocked_reason FROM tasks WHERE blocked_by IS NOT NULL AND status != 'done'"

# Counts
sqlite3 ~/claude-assistant/tasks.db \
  "SELECT (SELECT COUNT(*) FROM tasks WHERE status NOT IN ('done')) AS active, (SELECT COUNT(*) FROM tasks WHERE status='done') AS completed, (SELECT COUNT(*) FROM tasks WHERE blocked_by IS NOT NULL AND status != 'done') AS blocked"

sqlite3 ~/claude-assistant/tasks.db "INSERT INTO tasks (id, name, priority, problems, effort, status, subgoal, created_at, updated_at) VALUES ('T###', 'Name', 'P2', '1,4', 'M', 'pending', 'S1', '$(date +%Y-%m-%d)', '$(date -u +%Y-%m-%dT%H:%M:%SZ)')"

sqlite3 ~/claude-assistant/tasks.db "UPDATE tasks SET status='done', completed_at='$(date +%Y-%m-%d)', updated_at='$(date -u +%Y-%m-%dT%H:%M:%SZ)' WHERE id='T###'"

# Completed this week
sqlite3 ~/claude-assistant/tasks.db "SELECT COUNT(*) as done_this_week FROM tasks WHERE status='done' AND completed_at >= date('now', '-7 days')"

# Active count by priority
sqlite3 ~/claude-assistant/tasks.db "SELECT priority, COUNT(*) as count FROM tasks WHERE status != 'done' GROUP BY priority ORDER BY priority"

# Average days to complete
sqlite3 ~/claude-assistant/tasks.db "SELECT ROUND(AVG(julianday(completed_at) - julianday(created_at)), 1) as avg_days FROM tasks WHERE status='done' AND completed_at IS NOT NULL"

© mp-web3, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in skills/tasks of mp-web3/claude-starter-kit.

• SKILL.md
• LEARNINGS.md

Open the folder on GitHubat commit 546c72c