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 MartinOlivero/saas-builder --skill integrationsgit clone --depth 1 https://github.com/MartinOlivero/saas-builderWrote 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/martinolivero/saas-builder/integrations)<a href="https://agentmods.dev/skills/martinolivero/saas-builder/integrations"><img src="https://agentmods.dev/badge/skills/martinolivero/saas-builder/integrations/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/martinolivero/saas-builder/integrations"><img src="https://agentmods.dev/badge/skills/martinolivero/saas-builder/integrations.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 | $0.00220 | $0.01990 |
| Opus 5.5 | $0.00088 | $0.00796 |
| Sonnet 5.5 | $0.00044 | $0.00398 |
| Haiku 4.5 | $0.00022 | $0.00199 |
Grade A, and why
integrations 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 29d 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.
How it starts
The opening of the file, as written. The whole thing — 101 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Integrations
Your own code does what you wrote. A third party does what it does today, which may differ from its documentation, from yesterday, and from its sandbox. This skill designs for that.
Analogy: an integration is a business partner in another time zone who answers by post. Letters arrive twice, out of order, or not at all, and sometimes the letter says "done" when nothing was done. You don't run the business on the letters — you keep your own books and reconcile.
Discovery (max 3 questions, only if unknown)
- Direction — do you read from them, write to them, or both?
- Do they push events (webhooks), or do you pull?
- Is the connection per-product (one key of yours) or per-customer (each user connects their own account)?
Step 1 — Inbound: treat the webhook as a doorbell
Take only the id from the webhook, then fetch the object from their API with your own credentials. Never write the webhook body into your tables.
This one decision removes a family of problems: forged payloads, bodies re-serialized by a gateway (which breaks signature checks), events arriving out of order, and fields that are empty in the event but present in the API.
Around it:
- Verify the signature — and reject when it is missing. A handler that skips verification if no signature header arrives has no verification.
- Deduplicate by their event or message id, stored with a unique constraint. Deliveries repeat.
- Batched webhooks: one bad item must not stop the rest.
try/catchper item, persist before you enqueue, and neverreturnout of the loop when one entry belongs to an unknown account — the other tenants in the same request go silent. - Expect your own writes to come back. Change something through their API and the webhook reports it to you; without a guard it overwrites the state you just set.
- Never depend on the webhook alone. Build the pull path too — a "sync now" action that rings the same doorbell by hand, or a reconciliation job. Webhooks get registered and then never fire, and you find out from a customer.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 29d ago First seen · 101 lines · 220 tokens per session scan A 15722e6034f1
integrations is a skill published in the GitHub repository MartinOlivero/saas-builder (5 stars, last pushed 26d ago), licensed MIT. It adds 220 tokens to every session and 1,990 once invoked, about $0.0009 per session on Opus 5.5. 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 2026-10-08.
integrations compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| integrationsMartinOlivero/saas-builder | Skill | 5 | 26d ago | MIT | A | 220 | Claude Code |
| architecture-paradigm-microkernelathola/claude-night-market | Skill | 348 | yesterday | MIT | A | 32 | Claude Code |
| plugin-systemellmos-ai/skills | Skill | 9 | 4d ago | MIT | A | 27 | Claude Code, Codex |
| saas-migration-auditparisgroup-ai/imersao-ia-setup | Skill | 1 | 1mo ago | MIT | A | 90 | Claude Code |
| multi-tenancycamilooscargbaptista/cto-toolkit | Skill | 8 | 7mo ago | MIT | A | 22 | Claude Code |
| moai-ref-api-patternsmodu-ai/moai-adk | Skill | 1.3k | 4d ago | Apache-2.0 | A | 85 | Claude Code |
| flow-next-impl-reviewgmickel/flow-next | Skill | 717 | yesterday | MIT | A | 34 | Claude Code, Codex |