Skip to content

Infisical is the open-source platform for securing secrets, certificates, access, and AI Agents.

fixes

Adrian Kahali5402e7724ca6
diff --git a/docs/documentation/platform/agent-vault/guides/modal.mdx b/docs/documentation/platform/agent-vault/guides/modal.mdx
index 98a6d579f6..eb87f34032 100644
--- a/docs/documentation/platform/agent-vault/guides/modal.mdx
+++ b/docs/documentation/platform/agent-vault/guides/modal.mdx
@@ -41,7 +41,7 @@ Every sidecar runs the same proxy from one proxy token, so you set up the proxy
- A sandbox can only allow a hostname for HTTPS on port 443, and the proxy listens for plain HTTP on port 17323, so you'd have to allow the proxy by a fixed IPv4 address
- If the proxy is reachable from the internet, anyone who gets a session token can use it from anywhere until the session ends
- A sidecar is only reachable from its own sandbox, so a session token that leaks from the sandbox doesn't work anywhere else.
+ A sidecar is only reachable from its own sandbox, so nothing outside the sandbox can send requests through it.
</Accordion>
## Prerequisites
@@ -195,7 +195,7 @@ func buildImage(ctx context.Context, mc *modal.Client) (*modal.App, *modal.Image
## Step 4: Run an agent in a sandbox
-This function runs one prompt in a new sandbox. Replace `<access-bundle-name>` at the top with your access bundle's name:
+This function runs one prompt in a new sandbox. Replace `<access-bundle-name>` at the top with your access bundle's name. If you use EU Cloud or self-host Infisical, also replace `https://app.infisical.com` with the URL of your instance:
<CodeGroup>
@@ -301,10 +301,11 @@ def run_agent(prompt: str) -> str:
agent.stdin.write_eof()
agent.stdin.drain()
output = agent.stdout.read()
- agent.wait()
+ if agent.wait() != 0:
+ raise RuntimeError(f"The agent failed:\n{agent.stderr.read()}")
return output
finally:
- # Revoke first, so the token stops working even if terminating fails.
+ # Revoke first, so the session ends even if terminating fails.
try:
requests.post(
f"{INFISICAL_URL}/api/v1/agent-vault/sessions/{session['id']}/revoke",
@@ -412,10 +413,12 @@ export async function runAgent(prompt: string): Promise<string> {
await agent.stdin.writeText(`${session.token}\n${prompt}`);
await agent.stdin.close();
const output = await agent.stdout.readText();
- await agent.wait();
+ if ((await agent.wait()) !== 0) {
+ throw new Error(`The agent failed:\n${await agent.stderr.readText()}`);
+ }
return output;
} finally {
- // Revoke first, so the token stops working even if terminating fails.
+ // Revoke first, so the session ends even if terminating fails.
try {
await infisical(`/api/v1/agent-vault/sessions/${session.id}/revoke`, { token: accessToken });
} finally {
@@ -539,7 +542,7 @@ func runAgent(ctx context.Context, mc *modal.Client, app *modal.App, image *moda
var sandbox *modal.Sandbox
// Deferred calls run last-in, first-out, so the session is revoked first
- // and the token stops working even if terminating fails.
+ // and ends even if terminating fails.
defer func() {
if sandbox != nil {
_, terminateErr := sandbox.Terminate(context.Background(), nil)
@@ -606,9 +609,14 @@ func runAgent(ctx context.Context, mc *modal.Client, app *modal.App, image *moda
if err != nil {
return "", err
}
- if _, err := agent.Wait(ctx, nil); err != nil {
+ code, err := agent.Wait(ctx, nil)
+ if err != nil {
return "", err
}
+ if code != 0 {
+ stderr, _ := io.ReadAll(agent.Stderr)
+ return "", fmt.Errorf("the agent failed with exit code %d:\n%s", code, stderr)
+ }
return string(stdout), nil
}
```
@@ -645,7 +653,7 @@ console.log(await runAgent("Reply with the word OK."));
modal.close();
```
-```go Go
+```go Go expandable
// main.go
package main
@@ -682,7 +690,7 @@ func main() {
Claude Code replies, even though its sandbox has no Anthropic API key and can't reach the internet.
<Check>
- Each run now gets its own sandbox, proxy, and session. The only credential in the agent's container is a session token that only works from inside that sandbox, and the token stops working when the run ends.
+ Each run now gets its own sandbox, proxy, and session. The only credential in the agent's container is a session token, and your code revokes the session when the run ends.
</Check>
To check which requests the agent made, [set up session logs](/documentation/platform/agent-vault/session-logs#setting-up-session-logs), then [view the session's logs](/documentation/platform/agent-vault/sessions#view-session-logs) in Agent Vault. Requests that got a credential have the outcome **Brokered**.