registry / AI-agent control-surface risk / punchcard-presence

punchcard-presence

Punchcard Discord Rich Presence and local AI usage tracking for Claude Code and Codex

This report shows the evidence for punchcard-presence@1.0.8. 6 of 7 known versions are blocked or flagged. See all versions.

AI Security Review

scanned 2mo ago · by lpm-firewall-ai

The package’s global postinstall directly triggers persistent Claude Code hook injection and daemon startup. This meets the block policy for unconsented install-time mutation of a foreign AI-agent control surface.

Static reason

One or more suspicious static signals were detected.; source matched previously finalized malicious package; routed for review

Trigger

Global npm installation invokes postinstall.

Impact

Future Claude Code lifecycle events execute this package’s CLI; the package also reads Claude/Codex activity data.

Mechanism

postinstall-driven Claude hook injection and background persistence

Policy narrative

On global installation, postinstall runs `punchcard install` without user interaction. That path writes command hooks into the user’s Claude settings for prompt, stop, subagent, and session events, then starts detached background processes. Because the default configuration enables start-at-login, it also installs platform autostart persistence. The hook target is a foreign AI-agent configuration, making this an unconsented control-surface mutation.

AI rationale

Evidence

scripts/postinstall.cjsbin/cli.jslib/claude-hooks.jslib/state.jslib/paths.jslib/autostart.js

Decision evidence

public snapshot

AI identified this as a dangerous AI-agent capability at 99.0% confidence with low false-positive risk.

Evidence for policy risk

Global postinstall silently invokes the install command.
Install enables the daemon, installs Claude hooks, and starts detached processes.
Hooks overwrite the user’s Claude settings with command hooks for six agent events.
Default settings enable start-at-login, so install also reaches OS autostart setup.

Evidence against

No automatic external HTTP endpoint was found in the postinstall chain.
Metrics upload code is gated behind the separate More Metrics feature.

Behavioral surface

Source

ChildProcessCryptoDynamicRequireEnvironmentVarsFilesystemNetworkShell

Supply chain

HighEntropyStringsUrlStrings

ManifestNo manifest risk signals triggered.

scanned 24 file(s), 142 KB of source, external domains: app.punchcardai.workers.dev, github.com, registry.npmjs.org, www.apple.com

Source & flagged code

12 flagged

package.jsonView file

•scripts.postinstall = node scripts/postinstall.cjs

High

Install Time Lifecycle Scripts

Package defines install-time lifecycle scripts.

package.jsonView on unpkg

•scripts.postinstall = node scripts/postinstall.cjs

Medium

Ambiguous Install Lifecycle Script

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

package.jsonView on unpkg

bin/cli.jsView file

2import fs from "node:fs/promises"; L3: import { spawn } from "node:child_process"; L4: import path from "node:path";

High

Child Process

Package source references child process execution.

bin/cli.jsView on unpkg · L2

•matchType = token_shingles matchedPackage = punchcard-presence@1.0.6 matchedPath = bin/cli.js matchedIdentity = npm:cHVuY2hjYXJkLXByZXNlbmNl:1.0.6 similarity = 0.750 shingleOverlap = 36 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.

bin/cli.jsView on unpkg

lib/tray.jsView file

195await ensureDaemonRunning(paths); L196: child = spawn("powershell.exe", [ L197: "-NoProfile",

High

Shell

Package source references shell execution.

lib/tray.jsView on unpkg · L195

scripts/postinstall.cjsView file

1const { spawnSync } = require("node:child_process"); L2: const path = require("node:path");

Medium

Dynamic Require

Package source references dynamic require/import behavior.

scripts/postinstall.cjsView on unpkg · L1

lib/autostart.jsView file

3Trigger-reachable persistence chain: manifest.bin -> bin/cli.js -> lib/autostart.js L3: import path from "node:path"; L4: import { execFile } from "node:child_process"; L5: import { promisify } from "node:util"; ... L59: function darwinContext(options = {}) { L60: const home = options.home || os.homedir(); L61: const uid = options.uid ?? (typeof process.getuid === "function" ? process.getuid() : null); ... L72: L73: async function launchctl(arguments_, options = {}) { L74: const execute = options.execFile || execFileAsync; ... L87: export async function launchAgentRuntimeStatus(options = {}) { L88: if ((options.platform || process.platform) !== "darwin") return null; L89: const context = darwinContext(options);

High

Trigger Reachable Persistence

A manifest entrypoint or package-local install chain reaches persistence behavior.

lib/autostart.jsView on unpkg · L3

•matchType = normalized_sha256 matchedPackage = punchcard-presence@1.0.6 matchedPath = lib/autostart.js matchedIdentity = npm:cHVuY2hjYXJkLXByZXNlbmNl:1.0.6 similarity = 1.000 summary = normalized source hash matched finalized malicious source

High

Known Malware Source Similarity

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

lib/autostart.jsView on unpkg

3import path from "node:path"; L4: import { execFile } from "node:child_process"; L5: import { promisify } from "node:util"; ... L59: function darwinContext(options = {}) { L60: const home = options.home || os.homedir(); L61: const uid = options.uid ?? (typeof process.getuid === "function" ? process.getuid() : null); ... L72: L73: async function launchctl(arguments_, options = {}) { L74: const execute = options.execFile || execFileAsync; ... L87: export async function launchAgentRuntimeStatus(options = {}) { L88: if ((options.platform || process.platform) !== "darwin") return null; L89: const context = darwinContext(options);

lib/more-metrics.jsView file

2import path from "node:path"; L3: import { spawn } from "node:child_process"; L4: import { pathToFileURL } from "node:url"; ... L8: export const MORE_METRICS_VERSION = "1.0.1"; L9: export const MORE_METRICS_API_BASE_URL = "https://app.punchcardai.workers.dev"; L10: L11: function packageConfig(env = process.env) { L12: const packageName = String(

High

Same File Env Network Execution

A single source file combines environment access, network access, and code or shell execution; review context before blocking.

lib/more-metrics.jsView on unpkg · L2

•matchType = normalized_sha256 matchedPackage = punchcard-presence@1.0.6 matchedPath = lib/more-metrics.js matchedIdentity = npm:cHVuY2hjYXJkLXByZXNlbmNl:1.0.6 similarity = 1.000 summary = normalized source hash matched finalized malicious source

High

Known Malware Source Similarity

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

lib/more-metrics.jsView on unpkg

scripts/tray.ps1View file

•path = scripts/tray.ps1 kind = build_helper sizeBytes = 24892 magicHex = [redacted]

Medium

Ships Build Helper

Package ships non-JavaScript build or shell helper files.

scripts/tray.ps1View on unpkg

Source & flagged code

12 flagged · loading source

package.jsonView file

•scripts.postinstall = node scripts/postinstall.cjs

High

Install Time Lifecycle Scripts

Package defines install-time lifecycle scripts.

package.jsonView on unpkg

•scripts.postinstall = node scripts/postinstall.cjs

Medium

Ambiguous Install Lifecycle Script

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

package.jsonView on unpkg

bin/cli.jsView file

2import fs from "node:fs/promises"; L3: import { spawn } from "node:child_process"; L4: import path from "node:path";

High

Child Process

Package source references child process execution.

bin/cli.jsView on unpkg · L2

Findings

8 High7 Medium5 Low

HighInstall Time Lifecycle Scriptspackage.json

HighChild Processbin/cli.js

HighShelllib/tray.js

HighSame File Env Network Executionlib/more-metrics.js

HighTrigger Reachable Persistencelib/autostart.js

HighKnown Malware Source Similaritylib/autostart.js

HighKnown Malware Source Similaritylib/more-metrics.js

HighKnown Malware Source Similaritybin/cli.js

MediumAmbiguous Install Lifecycle Scriptpackage.json

MediumDynamic Requirescripts/postinstall.cjs

MediumNetwork

MediumEnvironment Vars

MediumInstall Persistencelib/autostart.js

Affected versions and remediation

This report applies to punchcard-presence@1.0.8.

The versions list below shows the verdict for each recorded version.

Avoid installing punchcard-presence@1.0.8. Remove it from direct dependencies and check your lockfile for transitive copies.
Choose an independently verified alternative or release. This report does not establish that other versions are safe.
If this version ran, investigate the affected machine and build environment. Rotate credentials it could access and rebuild from a trusted environment.

Evidence last updated: 2026-08-17 04:14:05.508Z (UTC).

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

Versions

7 known versions

1.0.8latestshownMaliciousSoft block: AI-agent control surfaceBlockPublishedAug 17, 2026, 4:06 AM 1.0.7MaliciousSoft block: AI-agent control surfaceBlockPublishedAug 15, 2026, 12:52 AM 1.0.6MaliciousSoft block: AI-agent control surfaceBlockPublishedAug 15, 2026, 12:24 AM 1.0.5MaliciousSoft block: AI-agent control surfaceBlockPublished

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Affected versions and remediation

Versions

Related packages

Other AI-agent control-surface risk packages

Runtime surface

Behavioral surface

Related packages

Other AI-agent control-surface risk packages

Source & flagged code