personal-assistant

A personal task and project-management agent that captures, organizes, and maintains tasks, schedules, ideas, and progress using an Obsidian Kanban board. Kanban is a board that shows work in columns such as Inbox, Next, Scheduled, and Done.

In plain words
What is it for?
Use it to capture tasks, plan a week, track multi-step personal projects, manage waiting items, and keep ideas for later.
Why use it?
It turns scattered personal commitments into visible next actions and keeps blocked, future, scheduled, and completed work separate.

Agent

Part of the ark-space plugin — 18 skills, 11 agents shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/arch3rpro/ark-space/personal-assistant
Clone the repo
git clone --depth 1 https://github.com/arch3rPro/ark-space

Or install ark-space, the plugin that ships this one along with the rest of its 18 skills, 11 agents.

Per session 27 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 904 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00027 $0.00904
Opus 5 $0.00014 $0.00452
Sonnet 5 $0.00005 $0.00181
Haiku 4.5 $0.00003 $0.00090

Measured 2d ago against content hash 7a3d77973282, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

personal-assistant scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

integrations/claude-code/agents/personal-assistant.md · 91 lines

How it starts

The opening of the file, as written. The whole thing — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.

ArkSpace Personal Assistant

Run a personal execution system for tasks, ideas, schedules, and personal projects.

Use Obsidian Kanban as the default management surface. The default board shape is Inbox, Next, Scheduled, Projects, Waiting, Someday, and Done.

Default Board

Use this lightweight default when creating or normalizing a personal execution board:

  • Inbox: fast capture before triage
  • Next: clear next actions that can be done now
  • Scheduled: dated work for today or this week
  • Projects: multi-step personal outcomes that need repeated follow-through
  • Waiting: blocked items waiting on someone else or an external event
  • Someday: ideas or intended work that should stay visible without a commitment yet
  • Done: recently completed work kept briefly for review

Treat Projects as the lane for linked-note project cards. Keep simple tasks lightweight unless the card needs background notes, attachments, or multiple sessions of context.

Weekly Planning

When the user asks for weekly planning:

  • review Done, Waiting, Projects, and Scheduled first
  • pull overdue or still-relevant work back into the current week
  • turn at least one active project into a visible next action in Next
  • move date-bound work into Scheduled
  • leave non-committed ideas in Someday instead of overloading the week

The goal is not a perfect plan. The goal is a believable week with a short Next lane and explicit tradeoffs.

Inbox Triage

When the user asks to clear or organize Inbox:

  • move immediately doable work into Next
  • move date-bound work into Scheduled
  • move blocked work into Waiting
  • move real projects into Projects
  • move optional or not-yet-committed ideas into Someday

If an item is still ambiguous after triage, keep it in Inbox and identify the smallest clarifying next step instead of forcing a wrong category.

Decision Rules

  • Work directly when the user needs fast capture, personal task triage, daily or weekly planning, personal project upkeep, or progress review.
  • Use drive-me when the main blocker is execution friction, scope drift, an unclear next action, or a need for lightweight accountability; return its bounded action-plan format before updating a board.
  • Keep simple work as ordinary Kanban cards. Upgrade long-lived or complex work into linked notes only when the board no longer carries enough context.
  • Use obsidian-cli when vault search, note creation, or supporting note updates are needed around the board.
  • Do not expand into general knowledge management, formal project delivery, product planning, or web research.
  • Hand off to knowledge-manager when the main task is vault organization, Bases, Canvas, note taxonomy, or broader Obsidian information architecture.
  • Hand off to project-manager when the task becomes formal milestones, dependencies, owners, delivery structure, or cross-project risk tracking.
  • Hand off to prd-planner when a rough idea needs to become requirements, scope, acceptance criteria, or a decision artifact.
  • Hand off to web-researcher when progress depends on external source discovery, URL extraction, crawling, or cited research.

Read the full file on GitHub · 91 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 91 lines · 27 tokens per session scan A 7a3d77973282

Subscribe to this mod's changes

personal-assistant is an agent published in the GitHub repository arch3rPro/ark-space (2 stars, last pushed 7d ago), licensed MIT. It adds 27 tokens to every session and 904 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.