@respira/wordpress-mcp-server@8.6.0
WordPress MCP server + companion plugin giving AI native access to page builder content, not just the REST API. 252 tools, 356 with WooCommerce, 18 page builders and the Site Editor. Element-level edits, full page creation, HTML to builder conversion, sna
AI Security Review
scanned 8d ago · by lpm-firewall-aiThe bundled audit skills direct silent transmission of site-identifying telemetry and operational findings to the publisher’s service. The behavior is tied to explicit skill use, not an npm install hook, so warn rather than block.
Static reason
No blocking static signals were detected.; previous stored version diff introduced dangerous source
Trigger
An agent runs one of the bundled audit skills and completes its report.
Impact
This discloses site identity and operational findings to Respira; the payload may also include user and agent identifiers.
Mechanism
The skill instructions construct a JSON payload and send it in a fire-and-forget POST to Respira’s telemetry endpoint without a separate telemetry confirmation.
AI rationale
The bundled audit skills direct silent transmission of site-identifying telemetry and operational findings to the publisher’s service. The behavior is tied to explicit skill use, not an npm install hook, so warn rather than block.
Evidence
dist/index.jsdist/install-skills.jsskills/mobile-experience-report/SKILL.mdskills/technical-debt-audit/SKILL.mdskills/woocommerce-health-check/SKILL.mdskills/mobile-experience-report/telemetry.tsskills/technical-debt-audit/telemetry.tsskills/woocommerce-health-check/telemetry.ts
Network endpoints1
www.respira.press/api/skills/track-usage
Decision evidence
public snapshotAI called this Suspicious at 94.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- Three bundled audit skills instruct the agent to silently send a report-time telemetry request that includes the WordPress site URL and software versions.
- The payloads include site-specific findings, such as mobile issues, unused plugins and database bloat, or active payment gateway counts.
- The skills send fire-and-forget POST requests to Respira’s telemetry endpoint; their helper code also includes agent identity metadata.
- The installer is a separate CLI command that copies the bundled skills into the user’s Claude skills directory; existing skills are skipped unless overwrite is requested.
- The package manifest lists no install lifecycle hook. Skill installation and setup are command-driven.
- The setup target selects client config paths, but its config write requires an explicit install command and token; the flagged config.js location is diagnostic text.
Evidence against
- The telemetry recipient is the package’s own Respira service, and the reported data is usage and summarized site findings rather than observed passwords or conversation transcripts.
- The package does not install these skills through an npm lifecycle hook; the agent configuration setup is also explicitly invoked.
Source & flagged code
6 flagged · loading sourcedist/install-skills.jsView file
•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 unpkgdist/chat-hygiene.js#virtual:normalized:round1View file
•stage = ast_semantic_analysis; reason = ast_parse_error; limitedFiles = 4
High
Findings
3 High5 Medium4 Low
HighTrigger Reachable External Ai Agent Control Surface Mutationdist/install-skills.js
HighSemantic Analysis Limiteddist/chat-hygiene.js#virtual:normalized:round1
HighPrevious Version Dangerous Deltadist/server.js
MediumNetwork
MediumEnvironment Vars
MediumAi Review Evidenceskills/mobile-experience-report/SKILL.md
MediumAi Review Evidenceskills/technical-debt-audit/SKILL.md
MediumAi Review Evidenceskills/woocommerce-health-check/SKILL.md
LowScripts Present
LowFilesystem
LowHigh Entropy Strings
LowUrl Strings
Affected versions and remediation
This report applies to @respira/wordpress-mcp-server@8.6.0.
See all versions of @respira/wordpress-mcp-server for other recorded verdicts.
- Review the evidence and your use of @respira/wordpress-mcp-server@8.6.0 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.