@rigxyz/cli@0.14.2

Portable, runnable agent workspaces

AI Security Review

scanned 22d ago · by lpm-firewall-ai

The inspected hook installs a first-party package-owned agent extension automatically, which the supplied policy classifies as warning-only lifecycle risk. Its scoped destination, collision guard, and opt-out do not establish malicious intent or justify blocking.

Static reason

High-risk behavior combination matched malicious policy.; source matched previously finalized malicious package; routed for review; previous stored version diff introduced dangerous source

Trigger

npm installation runs the postinstall hook unless RIG_SKIP_SKILL_INSTALL is set to 1.

Impact

Agents can discover package-authored guidance, and later package installations can replace that managed guidance without a fresh prompt.

Mechanism

The hook writes the managed rig skill into existing .claude and .agents directories, preserving unrelated skills and skipping matching versions.

AI rationale

Evidence

package.jsonbin/postinstall.mjssrc/cli.mjs.claude/skills/rig/SKILL.md.agents/skills/rig/SKILL.md

Decision evidence

public snapshot

AI called this Suspicious at 94.0% confidence as Dangerous Capability with medium false-positive risk.

Evidence for warning

package.json automatically runs bin/postinstall.mjs after installation.
The hook writes a Rig-owned SKILL.md under agent skill directories, defaulting to the user's home directory.
Installation targets .claude and .agents without a consent prompt; RIG_SKIP_SKILL_INSTALL disables it.
Existing skills without the package ownership marker are preserved.
A matching skill version skips rewriting; subsequent installations can update the managed skill.
The explicit workspace installation command also writes project agent guidance and merges manifest permissions while preserving denies.

Evidence against

The automatic extension is confined to the package-owned rig skill rather than broad agent settings.
No concrete malicious payload execution, credential theft, or destructive install behavior was established.

Behavioral surface

Source

ChildProcessCryptoDynamicRequireEnvironmentVarsFilesystemNetworkShell

Source & flagged code

12 flagged

package.jsonView file

Lines 1-26json

2 "name": "@rigxyz/cli",

3 "version": "0.14.2",

4 "description": "Portable, runnable agent workspaces",

5 "type": "module",

6 "bin": {

7 "rig": "bin/rig.mjs"

8 },

9 "files": [

10 "bin/",

11 "src/",

12 "README.md"

13 ],

14 "scripts": {

15 "test": "node --test test/*.test.mjs",

16 "postinstall": "node bin/postinstall.mjs"

High

Install Time Lifecycle Scripts

Package defines install-time lifecycle scripts.

package.jsonView on unpkg · L16

Medium

Ambiguous Install Lifecycle Script

Install-time lifecycle script is not statically allowlisted and needs review.

package.jsonView on unpkg · L16

17 },

18 "engines": {

19 "node": ">=20"

20 },

21 "dependencies": {

22 "@clack/prompts": "^1.3.0",

23 "@rigxyz/tapd": "^0.6.0"

24 }

25}

src/collab.mjsView file

Lines 1-35javascript

1// rig <-> tap integration.

2//

3// Rig shells out to the `tapd` CLI and parses its versioned --json contract

4// (protocolVersion: 1). It never imports tapd as a library or touches the

5// state DB directly — with one exception: `rig attach` (see loadTapdLib()

6// below), which has no invite secret to hand a CLI verb and so reuses

7// tapd's writeBindingConfig/openStateDb primitives directly. See

8// ~/code/tap/rig-tap-integration-plan.md.

9//

10// Auth: rig owns identity. `rig login` writes a Clerk JWT to

src/cli.mjsView file

Lines 2306-2346javascript

2306 }

2307}

2308

2309async function runInstallCommands(config, cwd, { interactive, yes, runHooks }) {

2310 const statuses

bin/postinstall.mjsView file

Lines 1-28javascript

1#!/usr/bin/env node

2//

3// Post-install hook for @rigxyz/cli.

4//

5// Writes the identical rig SKILL.md to every harness's global skill

6// convention this CLI knows about, so it's known in any folder, not just

7// inside an initialised rig workspace, regardless of which agent CLI is

8// reading:

9// - ~/.claude/skills/rig/SKILL.md (Claude Code)

10// - ~/.agents/skills/rig/SKILL.md (Codex, Amp, and other harnesses that

src/hub.mjs#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.

src/hub.mjs#virtual:normalized:round1View on unpkg

src/files.mjsView file

•matchType = token_shingles matchedPackage = @rigxyz/cli@0.13.0 matchedPath = src/files.mjs matchedIdentity = npm:QHJpZ3h5ei9jbGk:0.13.0 similarity = 1.000 shingleOverlap = 48 summary = source token shingles overlapped finalized malicious source

High

Known Malware Source Similarity

Source file is highly similar to a previously finalized malicious package; route for source-aware review.

src/files.mjsView on unpkg

Source & flagged code

12 flagged · loading source

package.jsonView file

•scripts.postinstall = node bin/postinstall.mjs

High

Install Time Lifecycle Scripts

Package defines install-time lifecycle scripts.

package.jsonView on unpkg

•scripts.postinstall = node bin/postinstall.mjs

Medium

Ambiguous Install Lifecycle Script

Install-time lifecycle script is not statically allowlisted and needs review.

package.jsonView on unpkg

src/collab.mjsView file

•matchType = previous_version_dangerous_delta matchedPackage = @rigxyz/cli@0.14.1 matchedIdentity = npm:QHJpZ3h5ei9jbGk:0.14.1 similarity = 0.882 summary = stored previous version shares package body but lacks this dangerous source file

Critical

Previous Version Dangerous Delta

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

src/collab.mjsView on unpkg

Findings

2 Critical8 High5 Medium5 Low

CriticalAi Agent Control Hijackbin/postinstall.mjs

CriticalPrevious Version Dangerous Deltasrc/collab.mjs

HighInstall Time Lifecycle Scriptspackage.json

HighChild Processsrc/collab.mjs

HighShellsrc/cli.mjs

HighTrigger Reachable External Ai Agent Control Surface Mutationsrc/cli.mjs

HighRuntime Package Installsrc/collab.mjs

HighSemantic Analysis Limitedsrc/hub.mjs#virtual:normalized:round1

HighKnown Malware Source Similaritysrc/collab.mjs

HighKnown Malware Source Similaritysrc/files.mjs

MediumAmbiguous Install Lifecycle Scriptpackage.json

Medium

Affected versions and remediation

This report applies to @rigxyz/cli@0.14.2.

See all versions of @rigxyz/cli for other recorded verdicts.

Review the evidence and your use of @rigxyz/cli@0.14.2 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: 2026-09-30 19:49:23.366Z (UTC).

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