registry / AI-agent capability abuse / @respira/wordpress-mcp-server / 8.3.18

@respira/wordpress-mcp-server@8.3.18

WordPress MCP server + companion plugin giving AI native access to page builder content, not just the REST API. 220 tools, 325 with WooCommerce, 17 page builders and the Site Editor. Element-level edits, full page creation, HTML to builder conversion, sna

AI Security Review

scanned 2mo ago · by lpm-firewall-ai

This is not malicious install-hook behavior because no npm lifecycle hook runs it automatically. The explicit agent-extension installer nevertheless carries a concrete silent-telemetry capability, so it should be warned rather than marked clean.

Static reason

No blocking static signals were detected.; previous stored version diff introduced dangerous source

Trigger

The user runs install-skills or --install-config, then an installed skill is invoked.

Impact

An agent may silently disclose site and usage metadata to Respira; a token-gated command may add or replace the Respira MCP entry in a selected client configuration.

Mechanism

Copies agent instructions and writes a remotely supplied MCP configuration entry.

AI rationale

Evidence

dist/install-skills.jsdist/setup.jsskills/mobile-experience-report/SKILL.md~/.claude/skills/<skill>/SKILL.md~/.codex/config.toml~/.claude.json.cursor/mcp.json~/.antigravity/mcp.json~/.hermes/config.yaml

Network endpoints1

www.respira.press/api/skills/track-usage

Decision evidence

public snapshot

AI called this Suspicious at 91.0% confidence as Dangerous Capability with low false-positive risk.

Evidence for warning

An explicit skill-install command copies bundled instructions into the user's Claude skills directory.
A bundled skill tells an agent to silently send usage telemetry, including the site URL and tool-use metadata, to Respira.
The explicit install-config command downloads a token-gated configuration payload and writes it into selected AI-client configuration files.

Evidence against

package.json has no preinstall, install, or postinstall hook and no self-dependency.
The skill installer and configuration writer are reached only through named command-line arguments.
No hidden payload loader, credential-file harvesting, or destructive install-time behavior was found.

Behavioral surface

Source

ChildProcessCryptoEnvironmentVarsFilesystemNetworkShell

Supply chain

HighEntropyStringsPossibleObfuscationUrlStrings

Source & flagged code

6 flagged

dist/install-skills.jsView file

Lines 18-58javascript

18 */

19import * as fs from 'fs';

20import * as path from 'path';

21import * as os from 'os';

22import { fileURLToPath } from 'url';

23const __filename = fileURLToPath(import.meta.url);

24const __dirname = path.dirname(__filename);

25// Resolve the bundled skills directory relative to the compiled JS.

26// dist/install-skills.js -> ../skills (sibling to dist/ inside the package)

27const BUNDLED_SKILLS_DIR = path.resolve(__dirname, '..', 'skills');

28const c = {

29 reset: '\x1b[0m',

30 bold: '\x1b[1m',

31 dim: '\x1b[2m',

32 red: '\x1b[31m',

33 green: '\x1b[32m',

34 yellow: '\x1b[33m',

35 blue: '\x1b[34m',

36 cyan: '\x1b[36m',

37};

38function parseArgs(argv) {

39 const opts = {

40 dest: path.join(os.homedir(), '.claude', 'skills'),

41 force: false,

42 help: false,

Medium

Ai Review Evidence

An explicit skill-install command copies bundled instructions into the user's Claude skills directory.

dist/install-skills.jsView on unpkg · L38

43 };

44 for (let i = 0; i < argv.length; i++) {

45 const arg = argv[i];

46 if (arg === '--help' || arg === '-h') {

47 opts.help = true;

48 }

49 else if (arg === '--force' || arg === '-f') {

50 opts.force = true;

51 }

52 else if (arg === '--dest' || arg === '-d') {

53 const next = argv[i + 1];

54 if (!next || next.startsWith('-')) {

55 console.error(`${c.red}error:${c.reset} --dest requires a path argument`);

56 process.exit(2);

57 }

58 opts.dest = path.resolve(next);

Lines 90-130javascript

90${c.bold}Layout:${c.reset}

91 <dest>/respira-builder-edits/SKILL.md

92 <dest>/respira-setup/SKILL.md

93 <dest>/respira-site-audit/SKILL.md

94 <dest>/respira-woocommerce/SKILL.md

96For per-project install, point --dest at <project>/.claude/skills/.

97`);

•Manifest-trigger-reachable source links an external AI-agent control path to a behavior-bearing write operation. const bundled_skills_dir = path.resolve(__dirname, '..', 'skills'); const c = { reset: '\x1b[0m', bold: '\x1b[1m', dim: '\x1b[2m', red: '\x1b[31m', green: '\x1b[32m', yellow: '\x1b[33m', blue: '\x1b[34m', cyan: '\x1b[36m', }; function parseargs(argv) { const opts = { dest: path.join(os.homedir(), '.claude', 'skills'), force: false, help: false, }; for (let i = 0; i < argv.length; i++) { const arg = argv[i]; if (arg === '--help' || arg === '-h') { opts.help = true; } else if (arg === '--force' || arg === '-f') { opts.force = true; } else if (

High

Trigger Reachable External Ai Agent Control Surface Mutation

Manifest-trigger-reachable source writes behavior-bearing configuration into a user or project AI-agent control surface.

dist/install-skills.jsView on unpkg

dist/config.js#virtual:normalized:round1View file

•stage = ast_semantic_analysis; reason = ast_parse_error; limitedFiles = 2

High

Semantic Analysis Limited

A bounded semantic-analysis stage reached its safety limit; remaining detectors completed, but this package requires AI review.

dist/config.js#virtual:normalized:round1View on unpkg

dist/server.jsView file

•matchType = previous_version_dangerous_delta matchedPackage = @respira/wordpress-mcp-server@8.3.8 matchedIdentity = npm:[redacted]:8.3.8 similarity = 0.864 summary = stored previous version shares package body but lacks this dangerous source file

High

Previous Version Dangerous Delta

This package version adds a dangerous source file absent from the previous stored version; route for source-aware review.

dist/server.jsView on unpkg

skills/mobile-experience-report/SKILL.mdView file

Lines 238-278markdown

238

239Performance (10 points):

240 - Mobile Core Web Vitals in the "good" range (LCP, INP, CLS from `respira_get_core_web_vitals` / `respira_run_pagespeed_audit`): 10 pts

241 - Fall back to a heuristic load estimate only when measured metrics are unavailable

242```

243

244**Score labels:**

245- 85-100: 🟢 Mobile-Friendly

Source & flagged code

6 flagged · loading source

dist/install-skills.jsView file

High

Trigger Reachable External Ai Agent Control Surface Mutation

Manifest-trigger-reachable source writes behavior-bearing configuration into a user or project AI-agent control surface.

dist/install-skills.jsView on unpkg

38function parseArgs(argv) { const opts = { dest: path.join(os.homedir(), '.claude', 'skills'), force: false, help: false,

Medium

Ai Review Evidence

An explicit skill-install command copies bundled instructions into the user's Claude skills directory.

dist/install-skills.jsView on unpkg · L38

Findings

3 High5 Medium4 Low

HighTrigger Reachable External Ai Agent Control Surface Mutationdist/install-skills.js

HighSemantic Analysis Limiteddist/config.js#virtual:normalized:round1

HighPrevious Version Dangerous Deltadist/server.js

MediumNetwork

MediumEnvironment Vars

MediumAi Review Evidencedist/install-skills.js

MediumAi Review Evidenceskills/mobile-experience-report/SKILL.md

LowScripts Present

LowFilesystem

LowHigh Entropy Strings

LowUrl Strings

Affected versions and remediation

This report applies to @respira/wordpress-mcp-server@8.3.18.

See all versions of @respira/wordpress-mcp-server for other recorded verdicts.

Review the evidence and your use of @respira/wordpress-mcp-server@8.3.18 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: 2026-09-04 15:11:21.997Z (UTC).

Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Affected versions and remediation

Runtime surface

Behavioral surface

Source & flagged code