blogger-publisher@1.3.3
AI Security Review
scanned 3mo ago · by lpm-firewall-aiSource inspection found no lifecycle execution, remote payload loading, shell execution, or unconsented foreign agent-control mutation. The explicit AI-agent scaffolding and bundled populated OAuth configuration create a real security concern.
Static reason
One or more suspicious static signals were detected.; source matched previously finalized malicious package; routed for review; previous stored version diff introduced dangerous source
Trigger
User runs the CLI interactively and selects `init`, or explicitly launches `mcp-blogger-server`.
Impact
A user-authorized setup can influence project AI agents and enable content creation, modification, or deletion through configured Blogger credentials; bundled credentials are an exposure risk.
Mechanism
Explicit agent-instruction scaffolding plus authenticated Blogger/MCP automation.
AI rationale
Source inspection found no lifecycle execution, remote payload loading, shell execution, or unconsented foreign agent-control mutation. The explicit AI-agent scaffolding and bundled populated OAuth configuration create a real security concern.
Evidence
package.jsonbin/cli.jsbin/mcp-server.jslib/tui.jslib/config.jslib/updater.jslib/uploaders/github.jslib/uploaders/imgbb.jslib/mcp/server.jsAGENT.md.envlib/mcp/tools/posts.jslib/mcp/tools/pages.jslib/mcp/tools/comments.js
Network endpoints4
registry.npmjs.org/blogger-publisher/latestapi.imgbb.com/1/uploadapi.github.com/repos/${repo}/contents/${gitPath}drive.google.com/thumbnail?id=${fileId}&sz=w1000
Decision evidence
public snapshotAI called this Suspicious at 90.0% confidence as Dangerous Capability with low false-positive risk.
Evidence for warning
- `.env` contains populated OAuth-related configuration values, including a refresh-token-length value.
- `lib/tui.js` copies package `AGENT.md` into the caller's project during explicit interactive `init`.
- `AGENT.md` instructs AI agents to use this package's publishing and MCP capabilities.
- `lib/mcp/tools/posts.js`, `pages.js`, and `comments.js` expose authenticated create/update/delete operations.
Evidence against
- `package.json` has no preinstall, install, postinstall, or prepare lifecycle hook.
- Entrypoints only run through declared CLI/MCP binaries; no import-time payload execution was found.
- Network calls are package-aligned: Blogger/Drive image publishing, selected image CDNs, GitHub uploads, and npm update checks.
- No child-process, eval/vm, dynamic code loading, native binaries, or credential-exfiltration path was found.
- Image uploads read explicitly referenced local image files; publisher blocks image paths outside the working directory.
Behavioral surface
CryptoEnvironmentVars
Source & flagged code
6 flagged · loading source.envView file
1patternName = blocked_file
severity = critical
matchedText = .env
redactedSecretContext =
secretLikeLines = 4
L1: BLOG_ID=<redacted:19 hex>
L2: CLIENT_ID=<redacted:72 token-like>
L3: CLIENT_SECRET=<redacted:35 token-like>
L6: REFRESH_TOKEN=<redacted:103 token-like>
Critical
lib/uploaders/github.jsView file
•matchType = normalized_sha256
matchedPackage = blogger-publisher@1.3.2
matchedPath = lib/uploaders/github.js
matchedIdentity = npm:YmxvZ2dlci1wdWJsaXNoZXI:1.3.2
similarity = 1.000
summary = normalized source hash matched finalized malicious source
High
Known Malware Source Similarity
Source file is highly similar to a previously finalized malicious package; route for source-aware review.
lib/uploaders/github.jsView on unpkg1const axios = require('axios');
L2: const fs = require('fs');
...
L6: async function uploadImageToGitHub(authClient, localFilePath) {
L7: const token = process.env.GITHUB_TOKEN;
L8: const repo = process.env.GITHUB_REPO; // format: username/repo
...
L23: const uniqueFileName = `${hash}-${fileName}`;
L24: const imageBase64 = fileBuffer.toString('base64');
L25:
Findings
1 Critical4 High2 Medium5 Low
CriticalCritical Secret.env
HighKnown Malware Source Similaritylib/uploaders/github.js
HighKnown Malware Source Similaritylib/uploaders/imgbb.js
HighKnown Malware Source Similaritylib/uploaders/drive.js
HighPrevious Version Dangerous Deltalib/publisher.js
MediumNetwork
MediumEnvironment Vars
LowScripts Present
LowWeak Cryptolib/uploaders/github.js
LowFilesystem
LowHigh Entropy Strings
LowUrl Strings
Affected versions and remediation
This report applies to blogger-publisher@1.3.3.
See all versions of blogger-publisher for other recorded verdicts.
- Review the evidence and your use of blogger-publisher@1.3.3 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.