Pay-per-call safety guards for AI agents: injection, tool-call, signing, secret, x402-trust.
Déduit des transports déclarés par cette annonce (streamable-http). Un client absent de cette liste n’est pas écarté pour autant — c’est simplement quelque chose que Forge ne peut pas confirmer.
La vérification confirme l’identité de l’éditeur (la propriété du dépôt), pas la sûreté du code. L’analyse de sécurité couvre les CVE connues et les scripts d’installation suspects.
Lu depuis un véritable échange MCP initialize → tools/list contre l’endpoint déclaré. Aucun outil n’a été invoqué — tools/list est l’appel d’introspection en lecture seule que le protocole prévoit pour cela. Cela reflète ce que le serveur annonçait à cet instant ; un endpoint hébergé n’est figé sur aucune version et peut changer sans préavis.
https://paygent.obsmetrics.com/mcp7 outils · 1624 mssecure-code-reviewSecurity review of a code snippet or diff. Returns structured findings (severity, CWE, location, remediation). [security; up to 75c/call]Security review of a code snippet or diff. Returns structured findings (severity, CWE, location, remediation). [security; up to 75c/call]
| Paramètre | Type | Description |
|---|---|---|
| code* | string | Source code or unified diff to review |
| language | string | Language hint, e.g. typescript, python |
| context | string | Optional context about the code |
pr-summaryTurn a git diff into a clear PR description or release notes. [dev-tools; up to 30c/call]Turn a git diff into a clear PR description or release notes. [dev-tools; up to 30c/call]
| Paramètre | Type | Description |
|---|---|---|
| diff* | string | Unified git diff to summarise |
| style | string | e.g. conventional, changelog, executive |
x402-trust-auditVet an x402 counterparty before settling USDC: scores the advertised payment requirements AND (when supplied) the EIP-3009 authorization you are about to sign. Returns a machine-enforceable trust verdict (per-entry scores, coverage-honest trustScore, spend-constraint + tamper-evident fingerprint) f…Vet an x402 counterparty before settling USDC: scores the advertised payment requirements AND (when supplied) the EIP-3009 authorization you are about to sign. Returns a machine-enforceable trust verdict (per-entry scores, coverage-honest trustScore, spend-constraint + tamper-evident fingerprint) f…
| Paramètre | Type | Description |
|---|---|---|
| paymentRequirements* | — | The x402 payment requirements from the counterparty: the 402 `accepts` array, or a single object. |
| selectedOptionIndex | number | Index in the accepts array the buyer intends to settle (default 0). The verdict is scoped to it. |
| paymentPayload | — | The UNSIGNED EIP-3009 authorization the buyer is about to sign: { authorization|message: {from,to,value,validAfter,validBefore,nonce}, domain: {name,version,ch… |
| expected | object | Optional caller expectations. |
| spendPolicy | object | Optional buyer spend policy to evaluate against and to pin facilitators. |
| endpointUrl | string | Resource URL being paid (context only; never fetched). |
| serverMetadata | — | Optional server metadata the caller already holds (context only; not fetched). |
| context | string | Optional free-form context. |
sign-guardPre-sign safety oracle for agent wallets: submit the transaction or EIP-712 message you are about to sign and get a machine-enforceable verdict. Decodes the calldata/typed-data, flags the drainer toolkit (unlimited approvals, setApprovalForAll, permit/permit2 + EIP-3009 to an unexpected party, tran…Pre-sign safety oracle for agent wallets: submit the transaction or EIP-712 message you are about to sign and get a machine-enforceable verdict. Decodes the calldata/typed-data, flags the drainer toolkit (unlimited approvals, setApprovalForAll, permit/permit2 + EIP-3009 to an unexpected party, tran…
| Paramètre | Type | Description |
|---|---|---|
| tx | object | An EVM transaction you are about to sign. |
| typedData | object | An EIP-712 message you are about to sign (the off-chain drainer surface: permit, Permit2, EIP-3009). { domain, types, primaryType, message }. |
| expected | object | Your stated intent. Supplying it lets the verdict BIND the action; only a fully bound, clean action is auto-sign-safe. For an allowance, you MUST supply maxAmo… |
| spendPolicy | object | Optional buyer spend policy (context only). |
| context | string | Optional free-form context. |
inject-guardUntrusted-content guardrail for agents: submit a blob of text you are about to feed to your own LLM (scraped web content, a tool result, another agent's message) and get a machine-enforceable verdict - is this a prompt-injection / jailbreak / data-exfiltration / tool-hijack attempt? Returns a risk…Untrusted-content guardrail for agents: submit a blob of text you are about to feed to your own LLM (scraped web content, a tool result, another agent's message) and get a machine-enforceable verdict - is this a prompt-injection / jailbreak / data-exfiltration / tool-hijack attempt? Returns a risk…
| Paramètre | Type | Description |
|---|---|---|
| content* | string | The untrusted text to scan before you feed it to your LLM. |
| context | string | Optional: where the content came from (url, tool name, sender) - context only. |
tool-call-guardPre-execution safety oracle for agent actions: submit the tool call you are about to run (shell, http, sql, file, code, env) plus your stated intent, and get a machine-enforceable verdict before you execute it. Decodes what the call does, flags the danger toolkit (rm -rf, reverse shell, curl|sh, SS…Pre-execution safety oracle for agent actions: submit the tool call you are about to run (shell, http, sql, file, code, env) plus your stated intent, and get a machine-enforceable verdict before you execute it. Decodes what the call does, flags the danger toolkit (rm -rf, reverse shell, curl|sh, SS…
| Paramètre | Type | Description |
|---|---|---|
| call* | object | The tool call you are about to execute. |
| intent | string | What this call is for (natural language). Used by the classifier for intent-mismatch. |
| expected | object | Machine-checkable constraints. Supplying them lets the verdict BIND the call; only a positively-scoped, satisfied call is auto-exec-safe. |
| context | string | Optional: where the task/input came from (untrusted source label). |
secret-scanLeaked-credential guardrail for agents: submit a blob you are about to commit, log, post, or hand to another tool (a diff, a config, an .env, an LLM output) and get a machine-enforceable verdict - does it contain a live secret? Detects cloud keys (AWS), VCS tokens (GitHub/GitLab), provider API keys…Leaked-credential guardrail for agents: submit a blob you are about to commit, log, post, or hand to another tool (a diff, a config, an .env, an LLM output) and get a machine-enforceable verdict - does it contain a live secret? Detects cloud keys (AWS), VCS tokens (GitHub/GitLab), provider API keys…
| Paramètre | Type | Description |
|---|---|---|
| content* | string | The text to scan for leaked secrets (diff, config, .env, log line, LLM output). |
7 outils sur 7 ont publié une description.
Les noms et descriptions d’outils sont écrits par l’éditeur et affichés tels quels, comme du texte inerte. Ce sont les chaînes qu’un client MCP transmet à un modèle, alors Forge y recherche des motifs d’injection de prompt — tout constat apparaît avec l’analyse de sécurité ci-dessus. « Privilégié » est une correspondance de mot-clé sur le nom de l’outil, pas un audit de ce qu’il fait : un nom anodin peut tout de même tout faire.
Pay-per-call safety guards for AI agents: injection, tool-call, signing, secret, x402-trust.
Les noms cliquables ouvrent l’index Forge de toutes les entrées observées exposant cet outil. Parcourir tous les outils indexés.
Cette entrée ne publie aucun paquet npm : Forge n'a donc pas d'arbre de dépendances pour elle. C'est une lacune de couverture — pas une affirmation qu'elle n'a aucune dépendance.