bies-guardian
Deterministic sealed verdicts on public claims and startup ideas (0-LLM claim-safety guardian).
我該用這個嗎
品質與安全性
根據工具定義與協定合規性的自動化分析。
上下文成本
這是每次將伺服器的工具載入模型上下文時所消耗的約略 token 數量。數量越高，可用於其他工作的注意力就越少。
安裝
一鍵安裝
將以下內容加入你的 `claude_desktop_config.json` 檔案：
{
"mcpServers": {
"bies-guardian": {
"url": "https://bies.io/mcp"
}
}
}遠端端點
https://bies.io/mcpstreamable-http它能做什麼
工具清單
工具（3）
🟢claim_safety_verdict(copy, lang, market)
Deterministic, sealed pre-publication verdict on a PUBLIC claim (marketing copy, product page, ad headline). Returns per-claim verdict (PASS / CONDITIONAL_PASS / STOP), severity, and a reproducible replay_hash. A 0-LLM rules engine grounded in KR 표시광고법 / US FTC claim categories — NOT legal advice, NOT market proof or launch approval. Call this before a public claim ships.
輸入結構描述
{
"type": "object",
"properties": {
"copy": {
"type": "string",
"description": "The exact public claim text you intend to publish."
},
"lang": {
"type": "string",
"enum": [
"ko",
"en"
],
"description": "Language of the claim. Default ko."
},
"market": {
"type": "string",
"enum": [
"KR",
"US"
],
"description": "Regulatory jurisdiction. Default follows lang (ko→KR, en→US)."
}
},
"required": [
"copy"
]
}🟢idea_rehearsal(idea, lang, market)
Deterministic decision-rehearsal verdict on a founder's BUSINESS IDEA (not a public claim). Returns a graded verdict (Conditional GO / Clear one critical gate / Confirm the headwind / Verify the basics / Redesign the structure / Firm up the inputs / Hold·Redesign) driven by the venture's structural signals, with a reproducible seal. Classification is retrieval-grounded; the verdict and seal are reproducible. NOT market proof, NOT investment advice. Use to pressure-test an idea before committing.
輸入結構描述
{
"type": "object",
"properties": {
"idea": {
"type": "string",
"description": "The business idea in the founder's own words — what you sell, to whom."
},
"lang": {
"type": "string",
"enum": [
"ko",
"en"
],
"description": "Language. Default ko."
},
"market": {
"type": "string",
"enum": [
"KR",
"US"
],
"description": "Market. Default follows lang (ko→KR, en→US)."
}
},
"required": [
"idea"
]
}🟢rejudge_claims(artifact, observations, lang)
Re-judge a prior claim-safety verdict against new EVIDENCE. Pass the 'artifact' returned by claim_safety_verdict plus 'observations' (what happened). The top-level verdict NEVER changes on evidence (conservative by design); what moves is 4 axes — now-sayable lines (behind your owner-gate), permanently issuer-only claims, claims still short on evidence, and new risks. Deterministic, sealed, stateless (nothing stored).
輸入結構描述
{
"type": "object",
"properties": {
"artifact": {
"type": "object",
"description": "The full artifact object returned by claim_safety_verdict (the prior verdict being re-judged)."
},
"observations": {
"type": "array",
"items": {
"type": "object"
},
"description": "Evidence events. Each: { outcome: 'paid_yes'|'written_pilot'|'observed_metric'|'rejection', targets: [{ ref_kind: 'claim_id'|'evidence_id', ref: '<id from the artifact>' }], rejection_reason_ref? }."
},
"lang": {
"type": "string",
"enum": [
"ko",
"en"
],
"description": "Language. Default ko."
}
},
"required": [
"artifact",
"observations"
]
}社群
證據