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.
npx skills add ryan-haver/fusebase-mcp --skill app-backendgit clone --depth 1 https://github.com/ryan-haver/fusebase-mcpWrote 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.
[](https://agentmods.dev/skills/ryan-haver/fusebase-mcp/app-backend)<a href="https://agentmods.dev/skills/ryan-haver/fusebase-mcp/app-backend"><img src="https://agentmods.dev/badge/skills/ryan-haver/fusebase-mcp/app-backend/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.
<a href="https://agentmods.dev/skills/ryan-haver/fusebase-mcp/app-backend"><img src="https://agentmods.dev/badge/skills/ryan-haver/fusebase-mcp/app-backend.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00092 | $0.12070 |
| Opus 5.5 | $0.00037 | $0.04828 |
| Sonnet 5.5 | $0.00018 | $0.02414 |
| Haiku 4.5 | $0.00009 | $0.01207 |
Grade A, and why
app-backend scanned grade A with 1 finding 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 3d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
When the app needs to receive webhooks from a third-party service (Asana, GitHub, Stripe, Monday, etc.), register the subscription with that provider **yourself** as part of the deploy — do not hand the user a `curl` / a This is a copy
97% identical to app-backend — 8 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 980 lines — stays where its author put it; the contents beside it link to each section on GitHub.
App Backend
Multi-User Architecture
Apps are always multi-user. The backend serves requests from many users concurrently. Every design decision must account for this.
Per-user vs. shared state:
| Storage | Scope | Use for |
|---|---|---|
| httpOnly cookies | Per-user (per browser) | OAuth tokens, user preferences, session data |
| Dashboard rows (keyed by user ID) | Per-user (persistent) | User settings, saved state |
| Fusebase secrets / env vars | Shared (all users) | API keys, service-account credentials |
| In-memory variables | Shared (all users, lost on restart) | Short-lived caches only |
Common mistakes:
- ❌ Storing a user's OAuth token in an env var or in-memory config → all users share one token
- ❌ Storing a user's preference in a module-level variable → last user's preference wins for everyone
- ❌ Using env vars for per-user settings or selections → same value for everyone
- ❌ Using session/token-derived values (
ft:*, JWT pieces, rotating token IDs) as persistent row partition keys - ✅ Store per-user data in cookies or dashboard rows keyed by user
- ✅ Use env vars only for credentials/config shared across all users (e.g. OAuth client ID/secret)
Shared live reads
A page many people sit on must not call the isolated store once per open browser. Read once per backend process and answer the waiting requests from that result. One user action is one runIsolatedStoreSqlBatch; do not run several store calls in parallel for it. An idle page does not poll every 1–2 seconds. Limits: fusebase-gate/references/isolated-sql.md (Connection budget).
Stable key as partition key (required)
For any persisted per-user data, derive the partition key from stable identity only:
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.
- 3d ago Changed · +4 lines ffaebccb1ab9
- 1mo ago First seen · 976 lines · 92 tokens per session scan A 283a6b315332
app-backend is a skill published in the GitHub repository ryan-haver/fusebase-mcp (8 stars, last pushed 3d ago), licensed MIT. It adds 92 tokens to every session and 12,070 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). It is 97% identical to app-backend, differing in 8 lines, and is treated as a copy.
app-backend compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| app-backendryan-haver/fusebase-mcp | Skill | 8 | 3d ago | MIT | A | 92 | Claude Code |
| mem0-oss-to-platformmem0ai/mem0 | Skill | 69k | yesterday | Apache-2.0 | A | 217 | Claude Code, Codex |
| azure-mgmt-botservice-dotnetmicrosoft/skills | Skill | 3.2k | yesterday | MIT | A | 78 | Claude Code |
| fastapi-router-pymicrosoft/skills | Skill | 3.2k | yesterday | MIT | A | 46 | Claude Code |
| dev-runtimelangwatch/langwatch | Skill | 5.0k | today | Apache-2.0 | A | 172 | Claude Code |
| api-designericrisco/rsc-harness | Skill | 529 | 3d ago | MIT | A | 105 | Claude Code, Codex |
| migrate-segw-to-raparc-mcp/arc-1 | Skill | 253 | today | MIT | A | 106 | Claude Code |