multichannel-secret-egress-v1
6 rules · Keryn Software Licence · rules 0.8.0 · all packs
How to read a rule
- Severity
- How bad the finding is if it is real:
errorfails a default gate,warningandinfodo not. - Confidence
- P0 exploitable as written; P1 dangerous in context; P2 a heuristic or a posture check: confirm by reading the code.
- False-positive risk
- low: an exact API or flag, so hits are almost always real; medium: the pattern is legitimate in many programs, so check each hit; high: it finds a lead to confirm, so expect benign hits.
- Edge
- The relationship the finding sits on. reaches: content or traffic gets to a thing; calls: one thing invokes another; reads: one thing reads from another; writes: one thing writes to another; sends: data leaves for another party; grants: authority is given or left unchecked; depends_on: something relies on a component it does not verify; trusts: one party believes another without evidence. The boundary it crosses, when there is one: user->prompt (untrusted input reaching a prompt); model->code (model output reaching code, a shell, a query or a page); secret->outside (a secret leaving the trusted side); supply-chain (code, models or prompts pulled in from outside).
- Taxonomy chips
- References for readers who work in OWASP LLM 2025, OWASP Agentic, CWE or NIST AI RMF. They say where a rule belongs, not that a framework is satisfied.
keryn.egress.py.secret-to-side-channel
Section titled “keryn.egress.py.secret-to-side-channel”error python data-flow rule python, requests, httpx
A secret (environment variable, config field, or credential-named attribute) reaches an outbound request URL/body or a raised exception without passing through a redactor. Filtering the user-facing response does not help if the secret still leaves through a tool call, an outbound request, or an error message. Redact on every channel, or never put the raw secret on one.
Confidence P1: dangerous in context. False-positive risk medium: the pattern is legitimate in many programs, so check each hit.
Edge inspected: sends data leaves for another party; it crosses the secret->outside boundary (a secret leaving the trusted side).
CWE-200 CWE-201 CWE-532 LLM02 LLM06 ASI03
Fix: Apply the same redaction to every outbound channel (tool-call arguments, request URLs and bodies, exception and error text), or keep the raw secret out of them entirely; send a credential only in the Authorization header of the service that owns it.
References: https://cwe.mitre.org/data/definitions/201.html · https://genai.owasp.org/llmrisk/llm022025-sensitive-information-disclosure/
Try it: a snippet that triggers this rule
Written for this page, not taken from any project. Save it as app/billing_sync.py and scan it with keryn scan app --verbose; the finding below appears.
import osimport requests
BASE = "https://billing.example.com/v2"
def fetch_invoices(customer_id: str): api_token = os.environ["BILLING_API_TOKEN"] url = f"{BASE}/customers/{customer_id}/invoices?token={api_token}" resp = requests.get(url, timeout=10) resp.raise_for_status() return resp.json()["items"]
if __name__ == "__main__": for inv in fetch_invoices("cus_1042"): print(inv["number"], inv["amount_due"])Checked with Keryn’s browser engine when this page was built: it reports keryn.egress.py.secret-to-side-channel at line 10 (error, proof 2). The rule’s own logic is not published, so it cannot run in this tab; the pack list in the playground shows what does.
keryn.egress.ts.secret-to-side-channel
Section titled “keryn.egress.ts.secret-to-side-channel”error typescript, javascript data-flow rule typescript, nodejs, axios
A secret (process.env or a credential-named field) reaches an outbound request URL/body or a thrown error without passing through a redactor. Filtering the user-facing response does not help if the secret still leaves through a tool call, an outbound request, or an error message. Redact on every channel, or never put the raw secret on one.
Confidence P1: dangerous in context. False-positive risk medium: the pattern is legitimate in many programs, so check each hit.
Edge inspected: sends data leaves for another party; it crosses the secret->outside boundary (a secret leaving the trusted side).
CWE-200 CWE-201 CWE-532 LLM02 LLM06 ASI03
Fix: Apply the same redaction to every outbound channel, or keep the raw secret out of them; send a credential only in the Authorization header of the service that owns it.
References: https://cwe.mitre.org/data/definitions/201.html · https://genai.owasp.org/llmrisk/llm022025-sensitive-information-disclosure/
Try it: a snippet that triggers this rule
Written for this page, not taken from any project. Save it as src/notify.js and scan it with keryn scan src --verbose; the finding below appears.
const webhook = "https://hooks.example.com/ops";
export async function reportFailure(job, err) { const key = process.env.PAYMENTS_API_KEY; const res = await fetch(`${webhook}?job=${encodeURIComponent(job)}&key=${key}`, { method: "POST", headers: { "content-type": "application/json" }, body: JSON.stringify({ job, message: err.message }), }); if (!res.ok) { throw new Error(`webhook returned ${res.status}`); }}Checked with Keryn’s browser engine when this page was built: it reports keryn.egress.ts.secret-to-side-channel at line 5 (error, proof 2). The rule’s own logic is not published, so it cannot run in this tab; the pack list in the playground shows what does.
keryn.egress.go.secret-to-side-channel
Section titled “keryn.egress.go.secret-to-side-channel”error go data-flow rule go
A secret (environment variable or credential-named field) reaches an outbound request URL or a formatted error without a redactor. Filtering the user-facing response does not help if the secret leaves through a request or an error. Redact on every channel, or keep the raw secret off them.
Confidence P1: dangerous in context. False-positive risk medium: the pattern is legitimate in many programs, so check each hit.
Edge inspected: sends data leaves for another party; it crosses the secret->outside boundary (a secret leaving the trusted side).
CWE-200 CWE-201 CWE-532 LLM02 LLM06 ASI03
Fix: Redact on every outbound channel, or keep the raw secret out of request URLs/bodies and error text; send a credential only in the Authorization header of the service that owns it.
References: https://cwe.mitre.org/data/definitions/201.html · https://genai.owasp.org/llmrisk/llm022025-sensitive-information-disclosure/
keryn.egress.java.secret-to-side-channel
Section titled “keryn.egress.java.secret-to-side-channel”error java data-flow rule java
A secret (environment variable or credential-named field) reaches an outbound request URL or a thrown exception without a redactor. Filtering the user-facing response does not help if the secret leaves through a request or an error. Redact on every channel.
Confidence P1: dangerous in context. False-positive risk medium: the pattern is legitimate in many programs, so check each hit.
Edge inspected: sends data leaves for another party; it crosses the secret->outside boundary (a secret leaving the trusted side).
CWE-200 CWE-201 CWE-532 LLM02 LLM06 ASI03
Fix: Redact on every outbound channel, or keep the raw secret out of request URLs and exception text; send a credential only in the Authorization header of the service that owns it.
References: https://cwe.mitre.org/data/definitions/201.html · https://genai.owasp.org/llmrisk/llm022025-sensitive-information-disclosure/
keryn.egress.php.secret-to-side-channel
Section titled “keryn.egress.php.secret-to-side-channel”error php data-flow rule php
A secret (getenv or $_ENV) reaches an outbound request URL or a thrown exception without a redactor. Filtering the user-facing response does not help if the secret leaves through a request or an error. Redact on every channel.
Confidence P1: dangerous in context. False-positive risk medium: the pattern is legitimate in many programs, so check each hit.
Edge inspected: sends data leaves for another party; it crosses the secret->outside boundary (a secret leaving the trusted side).
CWE-200 CWE-201 CWE-532 LLM02 LLM06 ASI03
Fix: Redact on every outbound channel, or keep the raw secret out of request URLs and exception text; send a credential only in the Authorization header of the service that owns it.
References: https://cwe.mitre.org/data/definitions/201.html · https://genai.owasp.org/llmrisk/llm022025-sensitive-information-disclosure/
keryn.egress.cs.secret-to-side-channel
Section titled “keryn.egress.cs.secret-to-side-channel”error csharp data-flow rule csharp, dotnet
A secret (Environment.GetEnvironmentVariable or a credential-named field) reaches an outbound request URL or a thrown exception without a redactor. Filtering the user-facing response does not help if the secret leaves through a request or an error. Redact on every channel.
Confidence P1: dangerous in context. False-positive risk medium: the pattern is legitimate in many programs, so check each hit.
Edge inspected: sends data leaves for another party; it crosses the secret->outside boundary (a secret leaving the trusted side).
CWE-200 CWE-201 CWE-532 LLM02 LLM06 ASI03
Fix: Redact on every outbound channel, or keep the raw secret out of request URLs and exception text; send a credential only in the Authorization header of the service that owns it.
References: https://cwe.mitre.org/data/definitions/201.html · https://genai.owasp.org/llmrisk/llm022025-sensitive-information-disclosure/