Skills/npm/core

core@0.36.0

CAUTION· install only

Publisher claims: "User files for a SaaS, under one person's authority: list, share with an expiry, revoke." Observed in sandbox: no network, touches nothing outside the project, writes no memory.

Registry npm (MCP server) · listed in the official MCP RegistryPublisher filelayer · publishing for 75 daysInstalls 6,552Last detonated 28 min ago

Coverage

Bundled scripts ran with no arguments; no agent task scenario for this bundle.
Tested
2 of 2 bundled scripts run with no arguments, with decoy credentials in the environment · every file read, file written and host contacted by what ran
Not tested
use of the skill by an agent

What changed since 0.35.2

vet-skills diff core 0.35.2 0.36.0

This version reaches for something 0.35.2 did not: 1 new host (filelayer.invalid) · 4 new credential variables (DATABASE_URL, S3_ACCESS_KEY_ID, S3_SECRET_ACCESS_KEY, S3_SESSION_TOKEN). Compared from the two recorded runs, not from source.

New hosts
filelayer.invalid
New credential variables
DATABASE_URL, S3_ACCESS_KEY_ID, S3_SECRET_ACCESS_KEY, S3_SESSION_TOKEN

The guard can hold an update like this until someone approves it (runtime.on_change: hold). A change that only shows with a real credential, or on a path the run did not reach, is not in a trace and so not in this comparison.

Is [email protected] safe to install?

KillerSkills ran it in the KillerSkills honeypot sandbox (2 scenarios; bundled scripts ran with no arguments; no agent task scenario for this bundle) on 2026-11-20 and scored it 74/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Auto-update or dynamic fetch configuration; Auto-update or dynamic fetch configuration; HTML comment carrying imperative instructions to the agent. Summary: no network, touches nothing outside the project, writes no memory.

Behavior trace

Honeypot run · 1 s · 2 scenarios · 0 deviations
t+CategoryObserved action
0.0sprocessno install scripts in bundle
0.0sprocessbundled script run in the sandbox with no arguments
0.2sprocessbundled script run in the sandbox with no arguments · not named in the instructions
0.4sprocessexited 2: the first real file when the client restarted this process.
0.4sprocess2 of 2 bundled scripts run with no arguments
0.4sprocessscript backend runs setup only
0.4sprocessscript backend runs setup only
0.4sprocessno install scripts to re-run

Static findings

HIGH
Auto-update or dynamic fetch configurationdist/authz.js:447 · Auto-update or dynamic fetch configuration — agent may load new code at runtime* hot-reloaded or monkey-patched role table.Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
HIGH
Auto-update or dynamic fetch configurationsrc/authz.ts:884 · Auto-update or dynamic fetch configuration — agent may load new code at runtime* hot-reloaded or monkey-patched role table.Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packagedist/cli.d.ts:7 · npx invocation of non-well-known package — may execute untrusted code* npx filelayer schema statusFix: Pin versions and sources; vendor anything fetched at install time.
AST06
CRITICAL
HTML comment carrying imperative instructions to the agentdocs/MCP.md:14 · HTML comment carrying imperative instructions to the agentdoccheck: skip reason="it connects a server to stdio and never returns, and it imports the MCP SDK, which is an optional peer this checker does not install"Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
HIGH
Auto-update or dynamic fetch configurationdist/authz.d.ts:357 · Auto-update or dynamic fetch configuration — agent may load new code at runtime* hot-reloaded or monkey-patched role table.Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packagedist/cli.js:6 · npx invocation of non-well-known package — may execute untrusted code* npx filelayer doctorFix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packagedist/cli.d.ts:6 · npx invocation of non-well-known package — may execute untrusted code* npx filelayer doctorFix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
File permission modification commandsrc/storage.ts:892 · File permission modification command — agent changes file ownership/permissions// file stricter than asked. A mutation removing that chmod survived theFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
LOW
Authorization escalation languagedocs/CLI.md:24 · Authorization escalation language · informational: not an instruction filepretend to. A convincing permission check layered on root access is worse thanFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
MEDIUM
File type combined with upload actionsrc/simple.ts:361 · File type combined with upload action — agent may exfiltrate local data files* log", as schema.sql puts it). A filter that excluded them would answer "noFix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
MEDIUM
npx invocation of non-well-known packagedist/cli.js:7 · npx invocation of non-well-known package — may execute untrusted code* npx filelayer schema statusFix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
File type combined with upload actiondist/simple.js:217 · File type combined with upload action — agent may exfiltrate local data files* log", as schema.sql puts it). A filter that excluded them would answer "noFix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
MEDIUM
Process kill commandtest/listing.test.ts:262 · Process kill command — agent can terminate processes// A soft-deleted org. Deleting one used to kill memberships and leave grantsFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
document.cookie accesstest/tiers.test.ts:160 · document.cookie access — can steal session cookiesconst { id } = await fl.files.put(utf8('<script>alert(document.cookie)</script>'), {Fix: Replace encoded or assembled strings with plain literals and remove undeclared network calls.
AST07
MEDIUM
Process kill commanddist/filelayer.js:1674 · Process kill command — agent can terminate processes-- survives the suite, and a guard no test can kill is a guard whoseFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
File permission modification commanddist/storage.js:625 · File permission modification command — agent changes file ownership/permissions// file stricter than asked. A mutation removing that chmod survived theFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
LOW
Authorization escalation languageCHANGELOG.md:1468 · Authorization escalation language · informational: not an instruction filetop of root access is worse than no check, because it reports that somethingFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
MEDIUM
File type combined with upload actionopenapi.json:18 · File type combined with upload action — agent may exfiltrate local data files…. All read authorization lives in `authz.ts`. There is no RLS policy in\n`schema.sql`, so a client that queries the tables directly is not filtered by\nanything and writes no audit event. Filelayer composes with database-level\nenforcement…Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
MEDIUM
npx invocation of non-well-known packagedist/mcp-server.d.ts:10 · npx invocation of non-well-known package — may execute untrusted code* npx would otherwise not install:Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
LOW
Authorization escalation languagellms.txt:116 · Authorization escalation language · informational: not an instruction file… and enforces -- the CLI cannot enforce anything. A permission check layered on root access is worse than no check: it reports that something was authorized when nothing w…Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
LOW
Authorization escalation languageCHANGELOG.md:1044 · Authorization escalation language · informational: not an instruction fileconvincing check on top of root access that the same header refuses. Reading asFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
MEDIUM
npx invocation of non-well-known packagesrc/cli.ts:6 · npx invocation of non-well-known package — may execute untrusted code* npx filelayer doctorFix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packagesrc/mcp-server.ts:157 · npx invocation of non-well-known package — may execute untrusted code' npx -y -p @filelayer/core -p pg -p @modelcontextprotocol/sdk -p zod filelayer-mcp',Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
AWS access key pattern detectedtest/cli.test.ts:392 · AWS access key pattern detected — possible hardcoded or harvested key!r.out.includes('AKIAIOSFODNN7EXAMPLE'),Fix: Never read or enumerate credential locations; accept a documented environment variable instead.
AST03
MEDIUM
AWS access key pattern detectedtest/storage.test.ts:34 · AWS access key pattern detected — possible hardcoded or harvested keyconst AK = 'AKIAFILELAYERTEST000';Fix: Never read or enumerate credential locations; accept a documented environment variable instead.
AST03
MEDIUM
AWS access key pattern detectedtest/cli.test.ts:387 · AWS access key pattern detected — possible hardcoded or harvested key'AKIAIOSFODNN7EXAMPLE</Message></Error>',Fix: Never read or enumerate credential locations; accept a documented environment variable instead.
AST03
MEDIUM
npx invocation of non-well-known packagedist/mcp-server.js:10 · npx invocation of non-well-known package — may execute untrusted code* npx would otherwise not install:Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packagesrc/mcp-server.ts:10 · npx invocation of non-well-known package — may execute untrusted code* npx would otherwise not install:Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
Process kill commandsrc/filelayer.ts:2347 · Process kill command — agent can terminate processes-- survives the suite, and a guard no test can kill is a guard whoseFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
npx invocation of non-well-known packagedist/mcp-server.js:141 · npx invocation of non-well-known package — may execute untrusted code' npx -y -p @filelayer/core -p pg -p @modelcontextprotocol/sdk -p zod filelayer-mcp',Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packagesrc/cli.ts:7 · npx invocation of non-well-known package — may execute untrusted code* npx filelayer schema statusFix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
Access to sensitive file pathstest/cli.test.ts:62 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files{ env: { ...process.env, DATABASE_URL: undefined, ...env } },Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05

Cost to the agent

Instructions loaded
≈89 tok
1 file · 81 in bundle

Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.

Known vulnerabilities

Not checked against OSV yet; the weekly check has not reached this version.

What would make this Verified

6 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%
  1. Unclaimed publisher≈ +8 on the totalClaim filelayer: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
  2. Hidden instructions≈ +6 on the totalRemove HTML comments, zero-width characters and white-on-white text that carry instructions. Everything the agent is meant to read should be visible to the person reading the file.
  3. Auto-update or dynamic fetch configuration≈ +4 on the totalPin versions and sources; vendor anything fetched at install time.
  4. npx invocation of non-well-known package≈ +2 on the totalPin versions and sources; vendor anything fetched at install time.
  5. File permission modification command≈ +2 on the totalRequest the narrowest permission that works; replace wildcards with the exact paths or hosts.
  6. File type combined with upload action≈ +2 on the totalRemove the call, or declare the host in frontmatter and explain in the README what is sent and why.
Publish a new version to npm, then queue it here. Not your skill? Claim the handle to get verdict changes by email.Claim filelayerPropose a patch to the maintainer