Borrowing it
Nothing to install: this file belongs to strale-io/strale. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/strale-io/strale/main/.agents/skills/vendor-switch/SKILL.mdgit clone --depth 1 https://github.com/strale-io/straleWrote 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/strale-io/strale/vendor-switch)<a href="https://agentmods.dev/skills/strale-io/strale/vendor-switch"><img src="https://agentmods.dev/badge/skills/strale-io/strale/vendor-switch/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/strale-io/strale/vendor-switch"><img src="https://agentmods.dev/badge/skills/strale-io/strale/vendor-switch.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 | $0.00082 | $0.01578 |
| Opus 5.5 | $0.00033 | $0.00631 |
| Sonnet 5.5 | $0.00016 | $0.00316 |
| Haiku 4.5 | $0.00008 | $0.00158 |
Grade A, and why
vendor-switch 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 2mo 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 — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/vendor-switch — replace an upstream vendor without leaving copy ghosts
Vendor switches are the highest-drift operation we perform. The platform silently keeps working on the new vendor while the methodology page, learn guides, learnGuides.ts, ToS, llms.txt, and any vendor-named copy still cite the old one — for as long as nobody notices. The 2026-04-30 cert audit found "OpenSanctions" cited 3 days after the move to Dilisense; the buyer-facing methodology page lied first.
This skill is the closing checklist. Run it BEFORE you push the vendor-switch PR, not after.
When this skill applies
Triggered by any of:
- The user says "we're switching from X to Y" (sanctions, PEP, adverse media, UBO, KYB, payments-card, payments-x402, embeddings, headless browser, log sink, risk-narrative LLM).
- A pull request title contains "drop", "switch", "migrate from X to Y" on a compliance / payments / log-sink capability.
- A new DEC entry is being drafted that supersedes a prior vendor-choice decision.
Step 1 — Verify the executor really uses the new vendor
# For sanctions/PEP/adverse-media style caps:
grep -rE "(api\\.|baseURL|client\\.)" apps/api/src/capabilities/<slug>.ts
Read the executor file. Confirm the new vendor's API URL appears and the old vendor's URL doesn't. If the executor has a fallback (e.g. Serper fallback for adverse-media), document that explicitly — the marketing copy must mention both primary AND fallback.
If the old vendor's code path is still callable, STOP. Either delete it or wrap it behind a feature flag with a documented sunset date.
Step 2 — Update STATIC_FACTS.vendors
apps/api/src/lib/platform-facts.ts — the canonical map. Update the
vendors field under STATIC_FACTS. Run the unit test:
cd apps/api && npx vitest run src/lib/platform-facts.test.ts
The test pins vendors.sanctions === "Dilisense" (or whatever the
current vendor is). Update the test in lockstep — that's the gate that
catches accidental reversions.
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.
- 2mo ago First seen · 161 lines · 82 tokens per session scan A 5f7abf986dca
vendor-switch is a skill published in the GitHub repository strale-io/strale (5 stars, last pushed yesterday), licensed MIT. It adds 82 tokens to every session and 1,578 once invoked, about $0.0003 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 2026-08-31.
vendor-switch compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| vendor-switchstrale-io/strale | Skill | 5 | yesterday | MIT | A | 82 | Claude Code, Codex |
| openapi-invoicingopenapi/openapi-skills | Skill | 15 | 1mo ago | MIT | A | 72 | Claude Code |
| openapi-riskopenapi/openapi-skills | Skill | 15 | 1mo ago | MIT | A | 59 | Claude Code |
| verify-polish-companybartosz-kuc/skanfirmy-mcp | Skill | 0 | 19d ago | MIT | A | 134 | Claude Code, Codex |
| expense-review-policyopenai/openai-cookbook | Skill | 77k | yesterday | MIT | A | 17 | Claude Code, Codex |
| kyc-doc-parseanthropics/financial-services | Skill | 41k | 9d ago | Apache-2.0 | A | 49 | Claude Code |
| akf-trust-metadatasickn33/agentic-awesome-skills | Skill | 49k | yesterday | MIT | A | 60 | Claude Code |