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 agentmods add commands/straddleio/ai/sandbox-testgit clone --depth 1 https://github.com/straddleio/aiWhat 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 | $0.00007 | $0.01769 |
| Opus 5 | $0.00003 | $0.00885 |
| Sonnet 5 | $0.00001 | $0.00354 |
| Haiku 4.5 | $0.00001 | $0.00177 |
Grade A, and why
sandbox-test 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 2d 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 — 167 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Straddle Sandbox Testing
If the user provided a scenario argument, jump directly to that scenario. Otherwise, present the list of available scenarios and ask which one to run.
General Rules
- All testing happens in sandbox (
https://sandbox.straddle.com). Never use production for testing. - Use the MCP
straddlecode tool for live execution against the sandbox API. - Explain each API response as it comes back. Do not silently move to the next step.
- If a status is unexpected, use
/explain-statusto diagnose. - Generate unique
external_idvalues for each test run to avoid idempotency collisions. - Always include
Idempotency-Keyheaders on POST/PATCH requests.
Documentation Sources
Before walking through any scenario, search product docs for current sandbox simulation parameters:
- Pay by Bank sandbox guide: https://docs.straddle.com/guides/resources/sandbox-paybybank
- Platform sandbox guide: https://docs.straddle.com/guides/resources/sandbox-platform
Search product docs by POSTing to https://docs.straddle.com/mcp to find the current sandbox_outcomes parameter values and simulation mechanisms. Do not guess simulation parameters -- always verify against current documentation.
Available Scenarios
- payment-lifecycle -- Full charge from created through paid
- payout-flow -- Payout with funding timing explanation
- failures -- Simulated insufficient funds, closed account, risk review
- ach-returns -- Simulated ACH returns (R01, R02, R05, R07, R10, R29)
- funding -- Query funding events after settlement
- bridge -- Bridge token and paykey creation
- embed-onboarding -- Full Embed account onboarding flow
- webhooks -- Webhook testing with Svix
1. payment-lifecycle
Walk through the full charge lifecycle: customer creation through payment completion.
Steps:
- Create a customer with test identity data. Confirm status reaches
verified. - Connect a bank account via manual entry (routing:
110000000, account:000123456789, type:checking). Extract the paykey. - Create a charge for $100 (amount: 10000, currency: USD). Use
consent_type: "internet"and today's date aspayment_date. - Check charge status. Walk through each transition:
created->scheduled->pending->paid. - Explain what happened at each stage and what webhooks would fire.
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.
- 2d ago First seen · 167 lines · 7 tokens per session scan A 463f41603eb4
sandbox-test is a command published in the GitHub repository straddleio/ai (5 stars, last pushed 3mo ago), licensed MIT. It adds 7 tokens to every session and 1,769 once invoked, about $0.0000 per session on Opus 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 2026-08-31.
Other commands, from other repositories
git
Git operations with intelligent commit messages and workflow optimization.
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
constitution
Create or update the project constitution from interactive or provided principle inputs.