abacus
AGENTS.md77 starsChanged 9 months agoApache-2.0View on GitHub
Use pnpm: pnpm install, pnpm dev, pnpm cli <cmd> AI commits MUST include: Use your actual model name (e.g., Claude Sonnet 4, Claude Opus 4.5). Routes require session validation except /sign-in, /api/auth/, /api/cron/, /api/webhooks/*. Use db-migrate skill for schema changes. See .claude/skills/db-migrate/SKILL.md Normalize at write-time via normalizeModelName(): sonnet-4, opus-4.5 (T), haiku-3.5 (HT) Use write-tests skill. See .claude/skills/write-tests/SKILL.md Use ui-design skill. See .claude/skills/ui-design/SKILL.md Use write-tip skill. See .claude/skills/write-tip/SKILL.md Use write-docs skill. See .claude/skills/write-docs/SKILL.md
We split this AGENTS.md into the rules it gives an agent
7 rules. About 0 are common, 1 is distinctive, and 6 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 5
Use pnpm: pnpm install, pnpm dev, pnpm cli <cmd>
- Very specific: line 13
Use your actual model name (e.g., Claude Sonnet 4, Claude Opus 4.5).
- Very specific: line 38
Use <AppLink> for internal links (preserves ?days=N)
- Very specific: line 39
Use useTimeRange() hook to access/update time range
- Very specific: line 43
Use db-migrate skill for schema changes.
- Distinctive: line 61
No gamification (no progress bars, XP, "level up")
- Very specific: line 70
Use write-tests skill. See .claude/skills/write-tests/SKILL.md
≈ “Common” is an estimate: at least 1 in 5 of the 36,368 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Agent Instructions
Package Manager
Use pnpm: pnpm install, pnpm dev, pnpm cli <cmd>
Commit Attribution
AI commits MUST include:
Co-Authored-By: <model-name> <noreply@anthropic.com>Use your actual model name (e.g., Claude Sonnet 4, Claude Opus 4.5).
API Routes
Routes require session validation except /sign-in, /api/auth/*, /api/cron/*, /api/webhooks/*.
import { NextResponse } from 'next/server';
import { wrapRouteHandlerWithSentry } from '@sentry/nextjs';
import { getSession } from '@/lib/auth';
async function handler(request: Request) {
const session = await getSession();
if (!session) return NextResponse.json({ error: 'Unauthorized' }, { status: 401 });
// handler logic
}
export const GET = wrapRouteHandlerWithSentry(handler, {
method: 'GET',
parameterizedRoute: '/api/your-route',
});Navigation
- Use
<AppLink>for internal links (preserves?days=N) - Use
useTimeRange()hook to access/update time range
Database
Use db-migrate skill for schema changes. See .claude/skills/db-migrate/SKILL.md
CLI
| Command | Description |
pnpm cli stats | DB statistics |
pnpm cli sync --days 7 | Sync recent usage |
pnpm cli backfill anthropic --from YYYY-MM-DD --to YYYY-MM-DD | Backfill provider data |
pnpm cli mappings:sync | Sync API key mappings |
Model Names
Normalize at write-time via normalizeModelName(): sonnet-4, opus-4.5 (T), haiku-3.5 (HT)
Adoption Stages
- Positive framing only (no "Struggling"/"At Risk")
- No gamification (no progress bars, XP, "level up")
- Stages: Exploring → Building Momentum → In the Flow → Power User
- Based on intensity (avg tokens/day), not frequency
- Inactive = 30+ days (hide by default)
- Thresholds: Power User 3M+, In Flow 1M+, Building Momentum 250K+
- See
src/lib/adoption.ts
Testing
Use write-tests skill. See .claude/skills/write-tests/SKILL.md
pnpm test # Run all tests
pnpm test:watch # Watch modeFrontend & UI
Use ui-design skill. See .claude/skills/ui-design/SKILL.md
Tips & Guides
Use write-tip skill. See .claude/skills/write-tip/SKILL.md
Documentation
Use write-docs skill. See .claude/skills/write-docs/SKILL.md
| Code Change | Update Docs |
| CLI commands | docs/src/content/docs/cli/*.mdx |
| Environment variables | docs/src/content/docs/getting-started/environment-variables.mdx |
| Provider setup/behavior | docs/src/content/docs/providers/*.mdx |
| Vercel config | docs/src/content/docs/deployment/vercel.mdx |
| Project structure | docs/src/content/docs/development/architecture.mdx |
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 28,693 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository5 other files it gives its agents.
- db-migrateSkill
- ui-designSkill
- write-docsSkill
- write-testsSkill
- write-tipSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

