mcp-runrunit@2.0.0
MCP server for Runrun.it API - Tasks, Comments, Projects
AI Security Review
scanned 3d ago · by lpm-firewall-aiThe package’s Sentry error reporting can include tool arguments, and its sanitizer does not remove ordinary strings that may contain identifying paths. Sentry is documented as optional and configurable, so this is a runtime telemetry concern rather than evidence of unrelated credential theft.
Static reason
High-risk behavior combination matched malicious policy.; previous stored version diff introduced dangerous source
Trigger
An MCP tool call fails while Sentry is enabled and a SENTRY_DSN is configured.
Impact
User-provided tool data, potentially including local project paths, may be sent to the configured Sentry destination with an error report.
Mechanism
The error handler attaches the tool arguments to the Sentry event context. Sanitization redacts selected sensitive-key names but preserves ordinary string values.
AI rationale
The package’s Sentry error reporting can include tool arguments, and its sanitizer does not remove ordinary strings that may contain identifying paths. Sentry is documented as optional and configurable, so this is a runtime telemetry concern rather than evidence of unrelated credential theft.
Evidence
dist/index.jsREADME.md
Decision evidence
public snapshotAI called this Suspicious at 86.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- On tool errors, the package adds the complete tool arguments to the Sentry event context.
- The sanitizer redacts selected object keys and truncates long strings, but leaves ordinary string values—including possible project paths—intact.
- Sentry monitoring is documented as optional and can be disabled; its destination is supplied through SENTRY_DSN.
Evidence against
- Runrun.it credentials are sent to the Runrun.it API for its stated task and comment functions.
- GitHub and Bitbucket credentials are used by their respective repository-sharing tools against their providers.
- The README documents Sentry configuration and a setting to disable event sending.
Behavioral surface
ChildProcessCryptoDynamicRequireEnvironmentVarsEvalFilesystemNetwork
HighEntropyStringsUrlStrings
Source & flagged code
9 flaggeddist/index.jsView file
1225// validation function arguments
L1226: data: new codegen_1.Name("data"),
L1227: // data passed to validation function
...
L2252: id = normalizeId(id);
L2253: return resolver.resolve(baseId, id);
L2254: }
...
L3112: for (i = 0; i < input.length; i++) {
L3113: code = input[i].charCodeAt(0);
L3114: if (code === 48) {
...
L12164: var StdioServerTransport = class {
L12165: constructor(_stdin = process2.stdin, _stdout = process2.stdout) {
L12166: this._stdin = _stdin;
Critical
Credential Exfiltration
Source appears to send environment or credential material to an external endpoint.
dist/index.jsView on unpkg · L12251225Trigger-reachable chain: manifest.main -> dist/index.js
L1225: // validation function arguments
L1226: data: new codegen_1.Name("data"),
L1227: // data passed to validation function
...
L2252: id = normalizeId(id);
L2253: return resolver.resolve(baseId, id);
L2254: }
...
L3112: for (i = 0; i < input.length; i++) {
L3113: code = input[i].charCodeAt(0);
L3114: if (code === 48) {
...
L12164: var StdioServerTransport = class {
L12165: constructor(_stdin = process2.stdin, _stdout = process2.stdout) {
L12166: this._stdin = _stdin;
Critical
Trigger Reachable Dangerous Capability
A package entrypoint or install-time lifecycle script reaches a source file with blocking dangerous behavior.
dist/index.jsView on unpkg · L1225Findings
3 Critical3 High5 Medium7 Low
CriticalCredential Exfiltrationdist/index.js
CriticalTrigger Reachable Dangerous Capabilitydist/index.js
CriticalPrevious Version Dangerous Deltadist/adapters/driven/git-workspace.js
HighChild Processdist/index.js
HighCommand Output Exfiltrationdist/index.js
HighTrigger Reachable Credential Exfiltrationdist/index.js
MediumDynamic Requiredist/server.js
MediumNetwork
MediumEnvironment Vars
MediumShips Build Helpercursor-skills/platforms/github/format-commit-message/validate-commit-msg.sh
MediumStructural Risk Force Deep Review
LowNon Install Lifecycle Scripts
Affected versions and remediation
This report applies to mcp-runrunit@2.0.0.
See all versions of mcp-runrunit for other recorded verdicts.
- Review the evidence and your use of mcp-runrunit@2.0.0 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.