/api/v1/agent-memory/governance/decisionsRead the agent-memory governance decision ledger (admin)
Returns the append-only governance decisions recorded for an organization, most recent first, together with a by-verdict summary. The summary is the point of MONITOR mode: `summary.byKind.flag` is the would-have-blocked count an operator watches to gauge false-positive rate before promoting a policy from `monitor` to `enforce`. The breakdown is deliberately NOT constrained by the `kind` filter — the breakdown is the pivot, so filtering to one verdict must not hide the other three. Read-only; the ledger is append-only at the database. Admin-only, and the query is always pinned to the server-resolved org, so one org can never read another's decision chain. If the backing migration has not been applied the route returns an empty, well-formed page rather than an error.
Authentication
Send Authorization: Bearer YOUR_API_KEY on every request. Generate API keys at /dashboard/settings/api-keys.
Parameters
orgId in queryOrganization to scope the read to. Resolved in order: this query param, then a body field, then the org an authenticated `eg_*` API key is bound to. Required for session auth (a user may belong to several orgs); optional for a key already scoped to one org. 400 MISSING_ORG_ID when none of the three yields a value.
stringsessionKey in queryRestrict to a single agent session. Also narrows the by-verdict summary.
stringkind in queryRestrict the returned page to one verdict. Does NOT constrain `summary.byKind`. An unrecognised value is ignored rather than rejected.
stringlimit in queryPage size, clamped to [1, 200]. Non-integer or absent falls back to 50.
integeroffset in queryRows to skip. Non-integer or absent falls back to 0.
integerResponse
200 example
{
"success": true
}All status codes
Code samples
cURL
curl -X GET \ https://evalguard.ai/api/v1/agent-memory/governance/decisions \ -H "Authorization: Bearer $EVALGUARD_API_KEY"
TypeScript
// The TypeScript SDK (@evalguard/sdk) exposes TYPED methods — runEval,
// getEval, runSecurityScan, checkFirewall, … — not a generic request().
// For an arbitrary endpoint, call it directly:
const res = await fetch("https://evalguard.ai/api/v1/agent-memory/governance/decisions", {
method: "GET",
headers: { Authorization: `Bearer ${process.env.EVALGUARD_API_KEY}` },
});
console.log(res.status, await res.json());Python
# The Python SDK (pip install evalguardai) exposes TYPED methods on
# EvalGuardClient — run_eval, get_eval, … — not a generic request().
# For an arbitrary endpoint, call it directly:
import os
import requests
headers = {"Authorization": f"Bearer {os.environ['EVALGUARD_API_KEY']}"}
response = requests.request("GET", "https://evalguard.ai/api/v1/agent-memory/governance/decisions", headers=headers)
print(response.status_code, response.json())Go
package main
import (
"context"
"fmt"
"net/http"
"os"
)
func main() {
req, _ := http.NewRequestWithContext(context.Background(), "GET", "https://evalguard.ai/api/v1/agent-memory/governance/decisions", nil)
req.Header.Set("Authorization", "Bearer "+os.Getenv("EVALGUARD_API_KEY"))
resp, err := http.DefaultClient.Do(req)
if err != nil { panic(err) }
defer resp.Body.Close()
fmt.Println(resp.Status)
}