registry / @delorenj/pjangler / 1.5.4

@delorenj/pjangler@1.5.4

Project subsystem bootstrapper CLI

AI Security Review

scanned 21d ago · by lpm-firewall-ai

Inspected source establishes guarded first-party agent hook installation, which warrants a warning under the supplied policy. Scanner claims alone do not establish credential theft, hidden payload activation, or malicious intent.

Static reason

One or more suspicious static signals were detected.

Trigger

Applying enabled notebook setup or running notebook migration with --apply.

Impact

Later agent sessions can execute the installed notebook integration.

Mechanism

The CLI installs its notebook skill and invokes a Python projector to persist SessionStart and SessionEnd hooks.

AI rationale

Evidence

dist/index.jsdist/assets/project-notebook-skill/scripts/project-hooks.py.claude/settings.json

Decision evidence

public snapshot

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

Evidence for warning

The manifest has no automatic installation hooks; its lifecycle hook runs before publication.
Notebook setup requires apply mode, an enabled policy, and a binding that is not disabled.
The integration invokes a packaged Python installer targeting Claude settings.
Notebook migration returns a dry run unless apply is requested.
The Python installer projects package-owned hooks into existing settings.

Evidence against

No concrete covert exfiltration or malicious payload execution was established.
The flagged agent locations define skill search roots, and the flagged Python test uses temporary directories.

Behavioral surface

Source

ChildProcessCryptoEnvironmentVarsFilesystemNetwork

Supply chain

HighEntropyStringsUrlStrings

ManifestNo manifest risk signals triggered.

scanned 8 file(s), 1.87 MB of source, external domains: 127.0.0.1, api.trello.com, plane.delo.sh, trello.com

Source & flagged code

8 flagged

dist/index.jsView file

•path = dist/index.js kind = trigger_reachable_excluded_payload sizeBytes = 886620 triggerReachableSource = dist/index.js excludedPayload = .mise/scripts/versioning.sh

Critical

Trigger Reachable Excluded Directory Payload

A package entrypoint or install-reachable source explicitly loads and activates an executable-looking payload from a test, fixture, or hidden path.

dist/index.jsView on unpkg

121// src/project/registryClient.ts L122: import { spawnSync } from "node:child_process"; L123: import { homedir as homedir2 } from "node:os";

High

Child Process

Package source references child process execution.

dist/index.jsView on unpkg · L121

•Manifest-reachable source captures an API credential, sends it to a fixed unofficial gateway, and persists that redirection in agent or shell configuration. dist/index.js: CREDENTIAL_KEY = /(?:^|[_-])(?:password|passwd|api[_-]?key|access[_-]?token|refresh[_-]?token|client[_-]?secret|private[_-]?key|credential|authorization)(?:$|[_-])/iu; `no Plane API key for workspace "${workspace}": set ${planeApiKeyNames(workspace).join(" or ")}, or add one to ${ctx.env.HERMES_FLEET_ENV?.trim() || join11(ctx.home, ".hermes", "fl... `Plane returned ${response.status} for ${path}` + (credentials ? ` \u2014 the API key was rejected; check ${planeApiKeyNames(client.workspace).join(" / ")}` : response.status === 4... import { chmodSync as chmodSync2, closeSync, copyFileSync as copyFileSync2, existsSync as ex

High

Credential Redirect Persistence

Manifest-reachable source sends a prompted API credential to a fixed unofficial gateway and persists the redirection.

dist/index.jsView on unpkg

7290if (typeof bin !== "string" || !bin) throw new Error(`${manifestPath} declares no skillex bin`); L7291: return { bin: resolve4(dirname4(manifestPath), bin), version: String(manifest.version ?? "unknown") }; L7292: } ... L7298: } catch (error) { L7299: process.stderr.write(`pjangler install is missing its @delorenj/skillex dependency (${error instanceof Error ? error.message : String(error)}); reinstall @delorenj/pjangler L7300: `); ... L7303: return await new Promise((resolveOutcome) => { L7304: const child = spawn(process.execPath, [cli.bin, ...args], { stdio: "inherit" }); L7305: const relay = (signal) => {

High

Command Output Exfiltration

Source combines command execution, command-output handling, and outbound requests; review data flow before blocking.

dist/index.jsView on unpkg · L7290

7290Trigger-reachable command-output exfiltration chain: scripts.start -> dist/index.js L7290: if (typeof bin !== "string" || !bin) throw new Error(`${manifestPath} declares no skillex bin`); L7291: return { bin: resolve4(dirname4(manifestPath), bin), version: String(manifest.version ?? "unknown") }; L7292: } ... L7298: } catch (error) { L7299: process.stderr.write(`pjangler install is missing its @delorenj/skillex dependency (${error instanceof Error ? error.message : String(error)}); reinstall @delorenj/pjangler L7300: `); ... L7303: return await new Promise((resolveOutcome) => { L7304: const child = spawn(process.execPath, [cli.bin, ...args], { stdio: "inherit" }); L7305: const relay = (signal) => {

High

Trigger Reachable Command Output Exfiltration

A manifest entrypoint or package-local install chain reaches command-output exfiltration behavior.

dist/index.jsView on unpkg · L7290

scripts/install-project-registry.mjsView file

Lines 1-35javascript

1#!/usr/bin/env node

2import { existsSync, mkdirSync, writeFileSync } from 'node:fs';

3import { homedir, userInfo } from 'node:os';

4import { dirname, join, resolve } from 'node:path';

5import { fileURLToPath }

dist/assets/project-notebook-skill/tests/test_project_hooks.pyView file

•path = dist/assets/project-notebook-skill/tests/test_project_hooks.py kind = payload_in_excluded_dir sizeBytes = 38660 magicHex = [redacted]

High

Payload In Excluded Dir

Package hides binary, compressed, or executable-looking payloads in test/fixture/hidden paths.

dist/assets/project-notebook-skill/tests/test_project_hooks.pyView on unpkg

•path = dist/assets/project-notebook-skill/tests/test_project_hooks.py kind = build_helper sizeBytes = 38660 magicHex = [redacted]

Medium

Ships Build Helper

Package ships non-JavaScript build or shell helper files.

dist/assets/project-notebook-skill/tests/test_project_hooks.pyView on unpkg

Source & flagged code

8 flagged · loading source

dist/index.jsView file

•path = dist/index.js kind = trigger_reachable_excluded_payload sizeBytes = 886620 triggerReachableSource = dist/index.js excludedPayload = .mise/scripts/versioning.sh

Critical

Trigger Reachable Excluded Directory Payload

A package entrypoint or install-reachable source explicitly loads and activates an executable-looking payload from a test, fixture, or hidden path.

dist/index.jsView on unpkg

121// src/project/registryClient.ts L122: import { spawnSync } from "node:child_process"; L123: import { homedir as homedir2 } from "node:os";

High

Child Process

Package source references child process execution.

dist/index.jsView on unpkg · L121

Findings

1 Critical6 High4 Medium5 Low

CriticalTrigger Reachable Excluded Directory Payloaddist/index.js

HighChild Processdist/index.js

HighSame File Env Network Executionscripts/install-project-registry.mjs

HighCredential Redirect Persistencedist/index.js

HighCommand Output Exfiltrationdist/index.js

HighTrigger Reachable Command Output Exfiltrationdist/index.js

HighPayload In Excluded Dirdist/assets/project-notebook-skill/tests/test_project_hooks.py

MediumNetwork

MediumEnvironment Vars

MediumShips Build Helperdist/assets/project-notebook-skill/tests/test_project_hooks.py

MediumStructural Risk Force Deep Review

Low

Affected versions and remediation

This report applies to @delorenj/pjangler@1.5.4.

See all versions of @delorenj/pjangler for other recorded verdicts.

Review the evidence and your use of @delorenj/pjangler@1.5.4 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: 2026-10-02 00:39:26.427Z (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