Package evidence
@getmcpm/[email protected]
Llm Injection Payload: AI-agent-directed instruction adjacent to credential exfil — prompt-injection payload (Shai-Hulud / SANDWORM_MODE). imperative="Ignore all previous instructions" target=".ssh/id_rsa"
Trust signals
Why this verdict
PkgRadar discounts a release’s score when public reputation argues against novel malware. The verdict above already reflects these — the panel just explains what was applied.
- Weekly downloads
- 442
- Versions published
- 21
- First published
- Mar 2026
- Publisher
- GitHub ActionsTrusted automation · −70% score
Effective trust discount applied: −70% (max across signals — discounts don’t stack). New install-lifecycle deltas vs the previous release would clear the discount.
Recommended action
Block this updateStatic evidence trips multiple high-signal indicators. Quarantine the release until the publisher validates the change or you can rule out the indicators below.
Block this release in CIcurl · GitHub Actions
Fail the build when this package version is added or upgraded. Replace $PKGRADAR_TOKEN with a Pro / Team API key from your dashboard.
curl -fsS https://pkgradar.com/gate/npm \
-H "Authorization: Bearer $PKGRADAR_TOKEN" \
-H "Content-Type: application/json" \
-d '{"specs":["@getmcpm/[email protected]"],"fail_on":"high"}'GitHub Actions step:
- name: PkgRadar gate
run: |
curl -fsS https://pkgradar.com/gate/npm \
-H "Authorization: Bearer ${{ secrets.PKGRADAR_TOKEN }}" \
-H "Content-Type: application/json" \
-d '{"specs":["@getmcpm/[email protected]"],"fail_on":"high"}'Why flagged
What the scanner saw
Llm Injection Payload: AI-agent-directed instruction adjacent to credential exfil — prompt-injection payload (Shai-Hulud / SANDWORM_MODE). imperative="Ignore all previous instructions" target=".ssh/id_rsa"
Not observed: package install, lifecycle script execution, or sandbox execution. PkgRadar only inspects on-disk artifacts.
Availability ledger
available
Status history (1 event)
- new → available · risk high · score 13 · status changed
Evidence
Static findings
2 static · 0 from release diff · showing high-signal first.
| Severity | Kind | Path | Detail | Points |
|---|---|---|---|---|
| high | Llm Injection Payload | package/dist/runner-GA67S7C5.js | AI-agent-directed instruction adjacent to credential exfil — prompt-injection payload (Shai-Hulud / SANDWORM_MODE). imperative="Ignore all previous instructions" target=".ssh/id_rsa" | 40 |
Show all 2 findings (low-signal and informational)
| Severity | Kind | Path | Detail | Points |
|---|---|---|---|---|
| high | Llm Injection Payload | package/dist/runner-GA67S7C5.js | AI-agent-directed instruction adjacent to credential exfil — prompt-injection payload (Shai-Hulud / SANDWORM_MODE). imperative="Ignore all previous instructions" target=".ssh/id_rsa" | 40 |
| low | Credential file access | package/dist/runner-GA67S7C5.js | matched ".ssh/" | 5 |
Manifest
Package metadata
Scripts8
buildtsupcleanrm -rf dist coveragedevtsup --watchlinttsc --noEmittestvitest runtest:coveragevitest run --coveragetest:watchvitesttypechecktsc --noEmit
Dependencies11
@inquirer/prompts^8.5.2@modelcontextprotocol/sdk^1.29.0@types/proper-lockfile^4.1.4chalk^5.6.2cli-table3^0.6.5commander^15.0.0ora^9.4.0proper-lockfile^4.1.2semver^7.8.1yaml^2.9.0zod^4.4.3