@pawlogic/dl@0.40.0
AI Security Review
scanned 2d ago · by lpm-firewall-aiThe updater’s runtime reports include high-cardinality identifiers such as user, prompt, tool-call, and sandbox IDs, without a user-facing disclosure or opt-out found in the package. The evidence supports a telemetry warning rather than credential theft or malicious exfiltration.
Static reason
One or more suspicious static signals were detected.
Trigger
The hosted wrapper runs dl-npm-update.sh before dl invocations; TTL-hit reports are sampled.
Impact
The configured callback service can associate update activity with users, projects, prompts, tool calls, and sandboxes; the package does not show a user-facing disclosure or opt-out for this reporting.
Mechanism
The helper posts updater status and project, user, prompt, tool-call, and sandbox identifiers to CALLBACK_SERVER_URL using DL_PROXY_TOKEN.
AI rationale
The updater’s runtime reports include high-cardinality identifiers such as user, prompt, tool-call, and sandbox IDs, without a user-facing disclosure or opt-out found in the package. The evidence supports a telemetry warning rather than credential theft or malicious exfiltration.
Evidence
dl-npm-update.shdl.mjspackage.json/tmp/dl-npm-checkedat/tmp/dl-npm-update.lock/home/user/.local/lib/node_modules/@pawlogic/dl
Network endpoints1
${CALLBACK_SERVER_URL}/sandbox-rpc/dl/selfupdate-event
Decision evidence
public snapshotAI called this Suspicious at 91.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- The manifest exposes dl and ilands CLI binaries; the package includes a hosted updater helper.
- The updater collects project, mode, tier, user, prompt, tool-call, and sandbox identifiers, plus a short npm error tail on failure.
- The helper sends updater events to the runtime-configured CALLBACK_SERVER_URL endpoint using DL_PROXY_TOKEN.
- The hosted wrapper is described as running the updater before each dl invocation; some TTL-hit reports are sampled.
Evidence against
- The manifest declares no lifecycle scripts.
- The reported data is updater metadata and a bounded npm error tail; the reviewed reporting code does not send credentials or general project files.
- Reporting requires CALLBACK_SERVER_URL and DL_PROXY_TOKEN to be present.
Behavioral surface
ChildProcessEnvironmentVarsEvalFilesystemNetwork
HighEntropyStringsMinifiedTrivial
Source & flagged code
7 flaggeddl.mjsView file
1#!/usr/bin/env node
L2: import e,{readFileSync as t,existsSync as i,mkdirSync as n,writeFileSync as o,chmodSync as r,realpathSync as a,rmSync as s,cpSync as c,constants as l}from"node:fs";import d,{dirnam...
High
1#!/usr/bin/env node
L2: import e,{readFileSync as t,existsSync as i,mkdirSync as n,writeFileSync as o,chmodSync as r,realpathSync as a,rmSync as s,cpSync as c,constants as l}from"node:fs";import d,{dirnam...
High
Same File Env Network Execution
A single source file combines environment access, network access, and code or shell execution; review context before blocking.
dl.mjsView on unpkg · L11#!/usr/bin/env node
L2: import e,{readFileSync as t,existsSync as i,mkdirSync as n,writeFileSync as o,chmodSync as r,realpathSync as a,rmSync as s,cpSync as c,constants as l}from"node:fs";import d,{dirnam...
High
Command Output Exfiltration
Source combines command execution, command-output handling, and outbound requests; review data flow before blocking.
dl.mjsView on unpkg · L1Findings
5 High4 Medium5 Low
HighChild Processdl.mjs
HighSame File Env Network Executiondl.mjs
HighCommand Output Exfiltrationdl.mjs
HighSandbox Evasion Gated Capabilitydl.mjs
HighTrigger Reachable Command Output Exfiltrationdl.mjs
MediumNetwork
MediumEnvironment Vars
MediumShips Build Helperdl-npm-update.sh
MediumStructural Risk Force Deep Review
LowEvaldl.mjs
LowFilesystem
LowHigh Entropy Strings
LowUrl Strings
LowNo License
Affected versions and remediation
This report applies to @pawlogic/dl@0.40.0.
See all versions of @pawlogic/dl for other recorded verdicts.
- Review the evidence and your use of @pawlogic/dl@0.40.0 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.