briefGitHub
POST /api/session → one link

Your agent writes the document. You decide inside it.

Brief turns an agent's work into an interactive decision document: sections, diagrams, charts, code, and the questions it needs you to answer. The agent sends one JSON payload and gets back one link.

$npx skills add ENEmyr/brief
See what it renders
claude code
> review the rate limiter plan· read 2 files · ran k6 · wrote briefReviewed the plan. Four calls need you:brief.algoryth.me/s/k2p9x4Qm7Zr1Ta
brief.algoryth.me/s/k2p9x4Qm7Zr1Ta2/4
Rate limiting for the public API
p95
2.4 s
after
0.2%
D21 60 / min2 600 / min3 Per plan
back to the agent
Reply to the "Rate limiting for the public API" doc (session k2p9x4Qm7Zr1Ta) - my decisions: [D2] Requests per minute per key - Choice: 600 / min - Note: separate burst allowance for webhooks? [D3] What to key the limit on - Choice: API key; User ID

How it works

  1. 01 / agent

    The agent posts a payload

    One JSON body: a title, sections of blocks, and the questions it needs answered. The skill teaches the agent the schema, so the agent writes the payload instead of hand-rolling an HTML report.

  2. 02 / you

    You read the link and answer

    The reply is a link. Open it on any device to read the sections, expand a diagram, highlight a line and leave a note, and pick an answer on each decision card.

  3. 03 / agent

    The agent reads you back

    Your answers and annotations come back as a ready-to-paste prompt, and the whole document stays readable as markdown at /raw, so the agent never regenerates what it already wrote.

Wire it up

Two paths to the same link. Pick the one your agent can run.

The skill teaches the agent the schema, ships an offline validator, and travels with the repository. This is the path most agents want.

terminal
# Install once, in the project your agent works in.
npx skills add ENEmyr/brief

# The agent now knows the payload schema and validates
# its own document offline before it posts anything.
# You ask for the work; it hands back a link:
#
#   "Reviewed the migration. Two calls need you:
#    https://brief.algoryth.me/s/k2p9x4Qm7Zr1Ta"

# Pull the schema again when new block types ship.
npx skills update brief

What you get back

A reading workspace, not a report. The question you are answering always sits next to the part of the document it is about.

  1. 1Outline marks where your notes and open questions are.
  2. 2Terms explain themselves on hover or tap.
  3. 3Diagrams play step by step, zoom with buttons.
  4. 4Decisions link to their section, take an Other answer and a reason.
  5. 5⌘K jumps to any section, block or decision.

Blocks

BLOCKS.md
Text and calloutsp list note warn good imageProse with inline code, links, the three callout tones, and images.
Datatable compare stat coverage detailsTables, side-by-side comparisons, stat rows, coverage bars, and collapsible detail.
Diagramsseq state layers erd mermaidSequence, state, layered architecture, and ERDs, each zoomable and pannable.
Code and mathba code mathSyntax-highlighted code, before-and-after diffs, and KaTeX math.
Chartsbigo heatmap histogram scatter plot3dComplexity curves, heatmaps, histograms, scatter plots, and 3D surfaces.
Agent workwalkthrough diff trajectoryCode that draws its own flowchart, semantic diffs with a change map, and an auditable log of what the agent did.
DecisionsdecisionYes/no and multiple-choice questions the reader answers in place.

Small things that add up

keyboardNumber keys pick options, J and K move between sections, ⌘K finds anything.
phoneDecisions open as a bottom sheet. Every target is at least 44px.
protectedArchive with a password. The doc is encrypted on your device before it is stored.
/rawEvery brief reads back as markdown, so the agent never rewrites what it already wrote.
printPDF export prints light, with your answers summarized at the end.
light and darkFollows the system setting until you pick one.

Give your agent a better way to ask.

One command in the project your agent works in.
$npx skills add ENEmyr/brief
Read the docs