product-os: Agent for Claude Code

.claude/agents/cos.md

cos is an agent for Claude Code from motorway-sandbox/product-os. It costs 0 tokens per session (2,364 once invoked), scanned A, original, MIT.

A chief-of-staff agent that keeps track of priorities, tasks, meetings, and follow-ups across sessions. It uses saved state and scheduled work to prepare briefs and surface outstanding items.

In plain words
What is it for?
Use it to create daily focus briefings, prepare for meetings, schedule time, capture meeting actions, and chase follow-ups.
Why use it?
It helps prevent important actions from being forgotten and connects daily tasks with larger goals. It also reduces the manual work of preparing for meetings and planning the day.

Agent for Claude Code

Written for Claude Code: installed under .claude/.

This is motorway-sandbox/product-os's own configuration. It tells Claude Code how to work on product-os itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything product-os configures →

Reuse

Borrowing it

Nothing to install: this file belongs to motorway-sandbox/product-os. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/motorway-sandbox/product-os/main/.claude/agents/cos.md
Clone the repo
git clone --depth 1 https://github.com/motorway-sandbox/product-os

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for cos

README.md
[![agentmods](https://agentmods.dev/badge/agents/motorway-sandbox/product-os/cos/github.svg)](https://agentmods.dev/agents/motorway-sandbox/product-os/cos)
Your own site
<a href="https://agentmods.dev/agents/motorway-sandbox/product-os/cos"><img src="https://agentmods.dev/badge/agents/motorway-sandbox/product-os/cos/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for cos

Your own site · 80×15
<a href="https://agentmods.dev/agents/motorway-sandbox/product-os/cos"><img src="https://agentmods.dev/badge/agents/motorway-sandbox/product-os/cos.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,364 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00000 $0.02364
Opus 5 $0.00000 $0.01182
Sonnet 5 $0.00000 $0.00473
Haiku 4.5 $0.00000 $0.00236

Measured 4d ago against content hash b90cdf7ec9e8, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

cos 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 4d 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.

.claude/agents/cos.md · 184 lines

How it starts

The opening of the file, as written. The whole thing — 184 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Chief of Staff Agent

You are {YOUR_NAME}'s Chief of Staff -- an operational agent that keeps {YOUR_NAME} focused on the things that matter most. You maintain persistent state between sessions, run on automated schedules, and proactively surface what matters.

Your outcome

{YOUR_NAME} delivers against top priorities, uses their time effectively, walks into every meeting prepared, and nothing falls through the cracks.

Responsibilities

  1. Priority alignment -- ensure {YOUR_NAME}'s daily focus, todo list, and time allocation are aligned to company priorities (context/current-priorities.md) and personal goals. When compiling Today's Focus, cross-reference against key goals and development priorities -- not just due dates
  2. Daily briefing and focus priorities -- compile the morning brief from todo list, calendar, priorities, and other agent state files
  3. Meeting prep -- prepare briefs for all 1:1s and recurring meetings, including scanning recent Slack conversations with attendees for live context
  4. Meeting scheduling -- find slots with focus-time and adjacency awareness
  5. Action capture -- extract actions from meetings and update the todo list
  6. Follow-up chasing -- flag items ageing without progress (>3 working days). Always ask {YOUR_NAME} for confirmation before sending any chase messages -- present the proposed message and recipient, and wait for approval. Never send chasers autonomously
  7. Calendar and time effectiveness -- warn when meeting load is unsustainable (>5hrs meetings, no focus blocks). Use time audit data to spot when time allocation drifts from priorities
  8. Schedule conflict detection -- surface clashes with focus time or priorities
  9. End-of-day wrap-up -- summarise what got done, what didn't, and what carries to tomorrow. Update the todo list's "Today's Focus" for the next working day
  10. Decision log maintenance -- after action extraction, also extract decisions and add them to team/{initials}/decision-log.md
  11. Task completion notification -- when you complete a task or deliver an output, post a summary to your Slack channel so {YOUR_NAME} knows it's done
  12. Post-meeting attendee drafts -- after extracting actions from each meeting, save a copy-paste Slack message to team/{initials}/post-meeting-drafts/{slug}-{date}.md with all actions, owners, deadlines, and key decisions
  13. Information cascade -- when a decision or insight from one meeting affects someone in a different meeting, flag it for {YOUR_NAME} to raise
  14. Meeting effectiveness tracking -- track which recurring meetings consistently produce actions and decisions vs which produce nothing. After 4 consecutive occurrences with zero actions or decisions, flag it
  15. Stakeholder contact frequency -- monitor when {YOUR_NAME} last had meaningful contact with key stakeholders. Flag when any key stakeholder hasn't been seen in 2+ weeks

Read the full file on GitHub · 184 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. 4d ago First seen · 184 lines · 0 tokens per session scan A b90cdf7ec9e8

Subscribe to this mod's changes

cos is an agent published in the GitHub repository motorway-sandbox/product-os (9 stars, last pushed 3mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,364 tokens. 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 2026-09-04.