Loading npm security reports…

opencode-agent-skill@15.9.0: Clean npm security report (Allow) | LPM Firewall

registry / opencode-agent-skill / 15.9.0

opencode-agent-skill@15.9.0

Pi-native UES coding-agent runtime with adaptive model-specific ACI, deterministic rehydration, universal reversible tool output, policy visibility compilation, diverse verification, and measured skill/context intelligence

AI Security Review

scanned 2h ago · by lpm-firewall-ai

No attack was identified in the inspected sources. The flagged files show fixture generation, managed configuration cleanup, process execution through a caller-supplied interface, and bounded browser inspection.

Static reason

No blocking static signals were detected.; source matched previously finalized malicious package; routed for review

Trigger

No package install-time trigger was found; the manifest hook is for prepublish verification.

Impact

No confirmed attack surface was established from the inspected behavior.

Mechanism

The inspected excerpts do not show credential collection, exfiltration, or unconsented broad agent configuration changes.

AI rationale

The flagged source excerpts align with fixture generation, managed cleanup, and runtime utilities, while the manifest shows no npm install hook or self-dependency. No concrete malicious behavior was established.

Decision evidence

public snapshot

AI called this Clean at 94.0% confidence as Benign with low false-positive risk.

Evidence for AI clean decision

The manifest exposes command-line binaries and only a prepublish verification hook; it has no npm install lifecycle hook or runtime self-dependency.
The retrieval fixture generates a miniature monorepo containing its own AGENTS.md file.
The installer excerpt only removes a marked managed block from the global AGENTS.md file.

Source & flagged code

6 flagged · loading source

evals/long/graders/long-horizon.mjsView file

13const load = async (file) => L14: import(pathToFileURL(path.join(workspace, "src", file)).href + "?eval=" + Date.now() + "-" + Math.random()) L15:

Medium

Dynamic Require

Package source references dynamic require/import behavior.

evals/long/graders/long-horizon.mjsView on unpkg · L13

evals/polyglot/fixtures/polyglot-bench/python/tenant_auth.pyView file

•path = [redacted]-bench/python/tenant_auth.py kind = payload_in_excluded_dir sizeBytes = 123 magicHex = [redacted]

High

Payload In Excluded Dir

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

evals/polyglot/fixtures/polyglot-bench/python/tenant_auth.pyView on unpkg

•path = [redacted]-bench/python/tenant_auth.py kind = build_helper sizeBytes = 123 magicHex = [redacted]

Medium

Ships Build Helper

Package ships non-JavaScript build or shell helper files.

evals/polyglot/fixtures/polyglot-bench/python/tenant_auth.pyView on unpkg

lib/content-artifacts.mjs#virtual:normalized:round1View file

•stage = ast_semantic_analysis; reason = ast_parse_error; limitedFiles = 5

High

Semantic Analysis Limited

A bounded semantic-analysis stage reached its safety limit; remaining detectors completed, but this package requires AI review.

lib/content-artifacts.mjs#virtual:normalized:round1View on unpkg

lib/process-runner.mjsView file

•matchType = normalized_sha256 matchedPackage = opencode-agent-skill@11.0.0 matchedPath = lib/process-runner.mjs matchedIdentity = npm:b3BlbmNvZGUtYWdlbnQtc2tpbGw:11.0.0 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/process-runner.mjsView on unpkg

lib/browser-runtime.mjsView file

•matchType = normalized_sha256 matchedPackage = opencode-agent-skill@11.0.0 matchedPath = lib/browser-runtime.mjs matchedIdentity = npm:b3BlbmNvZGUtYWdlbnQtc2tpbGw:11.0.0 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/browser-runtime.mjsView on unpkg

Source & flagged code

6 flagged

evals/long/graders/long-horizon.mjsView file

Lines 1-33javascript

1import assert from "node:assert/strict"

2import path from "node:path"

3import { spawnSync } from "node:child_process"

4import { fileURLToPath, pathToFileURL } from "node:url"

5

6const workspace = process.env.UES_EVAL_WORKSPACE

7const task = process.env.UES_EVAL_TASK

8if (!workspace || !task) {

9 console.error("UES_EVAL_WORKSPACE and UES_EVAL_TASK are required")

10 process.exit(2)

11}

12

13const load = async (file) =>

14 import(pathToFileURL(path.join(workspace, "src", file)).href + "?eval=" + Date.now() + "-" + Math.random())

15

Medium

Dynamic Require

Package source references dynamic require/import behavior.

evals/long/graders/long-horizon.mjsView on unpkg · L13

16const throws = (fn, Type) => assert.throws(fn, Type)

17

18switch (task) {

19 case "long-checkout-integration": {

20 const { calculateTotal } = await load("money.mjs")

21 const { reserveAll } = await load("inventory.mjs")

22 const { applyCharge } = await load("payment.mjs")

23 const { checkout } = await load("checkout.mjs")

24

25 const lines = [

26 { sku: "a", unitPriceCents: 250, quantity: 2 },

27 { sku: "a", unitPriceCents: 100, quantity: 1 },

28 { sku: "b", unitPriceCents: 50, quantity: 3 },

29 ]

30 assert.equal(calculateTotal(lines), 750)

31 throws(() => calculateTotal("x"), TypeError)

32 throws(() => calculateTotal([{ sku: "a", unitPriceCents: 1.5, quantity: 1 }]), TypeError)

33 throws(() => calculateTotal([{ sku: "a", unitPriceCents: -1, quantity: 1 }]), RangeError)

evals/polyglot/fixtures/polyglot-bench/python/tenant_auth.pyView file

•path = [redacted]-bench/python/tenant_auth.py kind = payload_in_excluded_dir sizeBytes = 123 magicHex = [redacted]

High

Payload In Excluded Dir

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

evals/polyglot/fixtures/polyglot-bench/python/tenant_auth.pyView on unpkg

•path = [redacted]-bench/python/tenant_auth.py kind = build_helper sizeBytes = 123 magicHex = [redacted]

Medium

Ships Build Helper

Package ships non-JavaScript build or shell helper files.

evals/polyglot/fixtures/polyglot-bench/python/tenant_auth.pyView on unpkg

lib/content-artifacts.mjs#virtual:normalized:round1View file

•stage = ast_semantic_analysis; reason = ast_parse_error; limitedFiles = 5

High

Semantic Analysis Limited

A bounded semantic-analysis stage reached its safety limit; remaining detectors completed, but this package requires AI review.

lib/content-artifacts.mjs#virtual:normalized:round1View on unpkg

lib/process-runner.mjsView file

•matchType = normalized_sha256 matchedPackage = opencode-agent-skill@11.0.0 matchedPath = lib/process-runner.mjs matchedIdentity = npm:b3BlbmNvZGUtYWdlbnQtc2tpbGw:11.0.0 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/process-runner.mjsView on unpkg

lib/browser-runtime.mjsView file

•matchType = normalized_sha256 matchedPackage = opencode-agent-skill@11.0.0 matchedPath = lib/browser-runtime.mjs matchedIdentity = npm:b3BlbmNvZGUtYWdlbnQtc2tpbGw:11.0.0 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/browser-runtime.mjsView on unpkg