Skills/npm/token-optimizer-mcp

token-optimizer-mcp@7.11.0

CAUTION· startup only

Publisher claims: "91.5% fewer tokens on tool output on Headroom's own benchmark, where Headroom removes 61.4%. Separately, the same SWE-bench tasks resolved at 0.71x the cost of plain Claude Code. Plus memory across sessions and subscription quota pacing. Claude Code, Code" Observed in sandbox: no network, touches nothing outside the project, writes no memory.

Registry npm (MCP server)Publisher ooples · publishing for 381 daysInstalls 15,740Last detonated 1 h ago
What would fix thisWatchDispute verdict

Coverage

Startup only; authenticated tool behavior not tested.
Tested
install scripts, with decoy credentials in place · second launch with the clock 45 days ahead · server startup over stdio (initialize sent) · every file read, file written and host contacted by what ran
Not tested
the tool list (the server exited before answering) · what the tools do when called with real credentials

What this means for the label: Caution here covers what ran. The server's tools were not exercised with a real credential, so nothing about their behavior is asserted; the credential summary below says what to do with that.

Before you give it your token

It will ask for
TOKEN_OPTIMIZER_CACHE_DIRTOKEN_OPTIMIZER_DAEMON_IDLE_MSTOKEN_OPTIMIZER_HOMETOKEN_OPTIMIZER_LAUNCH_IMPORT_ONLYTOKEN_OPTIMIZER_MODETOKEN_OPTIMIZER_PROXYTOKEN_OPTIMIZER_PROXY_ACCOUNTINGTOKEN_OPTIMIZER_PROXY_CAPTURETOKEN_OPTIMIZER_PROXY_CORE_TOOLSTOKEN_OPTIMIZER_PROXY_DEFER_HOLDOUTTOKEN_OPTIMIZER_PROXY_DROP_THINKINGTOKEN_OPTIMIZER_PROXY_KEEP_TOOLSTOKEN_OPTIMIZER_PROXY_KEEPWARM_MARGIN_MSTOKEN_OPTIMIZER_PROXY_KNOWLEDGE_CHARSTOKEN_OPTIMIZER_PROXY_NULLTOKEN_OPTIMIZER_PROXY_SMALL_TOOL_CHARSTOKEN_OPTIMIZER_PROXY_SPILLTOKEN_OPTIMIZER_PROXY_UPSTREAMTOKEN_OPTIMIZER_RUNTIMETOKEN_OPTIMIZER_SETTINGSTOKEN_OPTIMIZER_SHELL_PROFILESTOKEN_OPTIMIZER_UPDATE_CHECKTOKEN_OPTIMIZER_VERSIONConfiguration, not credentials: CLAUDE_CONFIG_DIR, ComSpec, MCP_GLOBAL_PATH, npm_config_cache, SOCKET_PATH, ZDOTDIR
Where it talks to
no vendor host named in the codeAlso elsewhere.example, githubusercontent.com, claude.com, nodejs.org — not the publisher's domain; check what for before connecting.
What the honeypot saw
no credential files read · no network
What we could not test
the tool list (the server exited before answering) · what the tools do when called with real credentials

Connect it with a scoped credential, under the guard, and read the third-party hosts first. The code names elsewhere.example, githubusercontent.com, claude.com, nodejs.org as well as its own vendor. The guard injects TOKEN_OPTIMIZER_CACHE_DIR, TOKEN_OPTIMIZER_DAEMON_IDLE_MS, TOKEN_OPTIMIZER_HOME, TOKEN_OPTIMIZER_LAUNCH_IMPORT_ONLY, TOKEN_OPTIMIZER_MODE, TOKEN_OPTIMIZER_PROXY, TOKEN_OPTIMIZER_PROXY_ACCOUNTING, TOKEN_OPTIMIZER_PROXY_CAPTURE, TOKEN_OPTIMIZER_PROXY_CORE_TOOLS, TOKEN_OPTIMIZER_PROXY_DEFER_HOLDOUT, TOKEN_OPTIMIZER_PROXY_DROP_THINKING, TOKEN_OPTIMIZER_PROXY_KEEP_TOOLS, TOKEN_OPTIMIZER_PROXY_KEEPWARM_MARGIN_MS, TOKEN_OPTIMIZER_PROXY_KNOWLEDGE_CHARS, TOKEN_OPTIMIZER_PROXY_NULL, TOKEN_OPTIMIZER_PROXY_SMALL_TOOL_CHARS, TOKEN_OPTIMIZER_PROXY_SPILL, TOKEN_OPTIMIZER_PROXY_UPSTREAM, TOKEN_OPTIMIZER_RUNTIME, TOKEN_OPTIMIZER_SETTINGS, TOKEN_OPTIMIZER_SHELL_PROFILES, TOKEN_OPTIMIZER_UPDATE_CHECK, TOKEN_OPTIMIZER_VERSION only into requests to the host you bind (and its subdomains) and refuses the rest.

$vet-skills run --net declared -- npx -y @ooples/token-optimizer-mcp

Never paste a credential into a chat, a ticket or a report, including ours. How the guard holds it.

Is [email protected] safe to install?

KillerSkills ran it in the KillerSkills honeypot sandbox (3 scenarios; startup only; authenticated tool behavior not tested) on 2026-11-15 and scored it 70/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Access to sensitive file paths; Collapsed <details> block carrying instructions; Collapsed <details> block carrying instructions. Summary: no network, touches nothing outside the project, writes no memory.

Behavior trace

Honeypot run · 49 s · 3 scenarios · 0 deviations
t+CategoryObserved action
0.0sprocessinstall script run in the sandbox
0.1sprocessexited 1: Node.js v22.23.5
29sprocessdependency install exit 0
49sclaimedstartup only: exited (0) before answering · code: 'ERR_MODULE_NOT_FOUND', url: 'file:///tmp/ks-detonate-rWKLQm/task/home/work/acme-api/.claude/skills/token-optimizer-mcp/dist/server/cli-dispatch.js' } Node.js v22.23.5
49sprocessscript backend runs setup only
49sprocesssecond launch, clock +45 days, CI unset
49sprocessexited 1: Node.js v22.23.5

Static findings

HIGH
Access to sensitive file pathsdist/rollout/cli.js:165 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesconst seeded = { ...env };Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Collapsed <details> block carrying instructionsREADME.md:71 · Collapsed <details> block carrying instructions<summary><strong>Requirements and what happens without them</strong></summary> - **Node.js 22.18.0+ on Node 22, or 24.11.0+**, for the hooks and the MCP server.Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
HIGH
Collapsed <details> block carrying instructionsREADME.md:230 · Collapsed <details> block carrying instructions<summary><strong>The knowledge graph</strong></summary> ``` you touch src/auth.ts │ ├─ verify() compares exp against the LOCAL clock (finding, 0.9) ├─ per-host Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
HIGH
Privileged sudo commanddist/learn/analyze.js:112 · Privileged sudo command — agent may execute with elevated privileges* discarded the rest of `sudo -E env FOO=1 python script.py` and left nothing thisFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
HIGH
Credential pass-through configurationdist/proxy/server.js:1603 · Credential pass-through configuration — agent relays sensitive credentialsthrow new Error(`token-optimizer proxy: refusing to forward credentials to ${upstream} -- ` +Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST02
HIGH
Collapsed <details> block carrying instructionsREADME.md:326 · Collapsed <details> block carrying instructions<summary><strong>Verify the release, or remove it</strong></summary> Releases are published from CI with npm provenance, so `npm audit signatures` ties the packFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
HIGH
Access to sensitive file pathspackage.json:60 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"!**/.env",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Collapsed <details> block carrying instructionsREADME.md:307 · Collapsed <details> block carrying instructions<summary><strong>Privacy and telemetry</strong></summary> - Nothing is measured about you unless you ask. Anonymous usage data takes two separate opt-ins: `TOKEFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
MEDIUM
Process kill commanddist/proxy/cli.js:262 · Process kill command — agent can terminate processes// ANSWERING A QUESTION IS NOT RUNNING, so this sits above the kill switch andFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
HIGH
Access to sensitive file pathspackage.json:61 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"!**/.env.*"Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Process kill commanddist/proxy/enabled.js:12 · Process kill command — agent can terminate processes/** Enabled by default; explicit opt-outs and the global kill switch win. */Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
npx invocation of non-well-known packagesrc/plugin/launch.mjs:8 · npx invocation of non-well-known package — may execute untrusted code* for why pinning was rejected), but it pays a per-launch cost that npx cannotFix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packagedist/update/cli.d.ts:11 · npx invocation of non-well-known package — may execute untrusted code* an npx cache entry, a Claude Code plugin, a source checkout, and an installFix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
Process kill commanddist/proxy/server.js:2122 · Process kill command — agent can terminate processes// used to kill immediately, so a clean stop dropped the tail of the ledgerFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
npx invocation of non-well-known packagedist/update/apply.js:10 · npx invocation of non-well-known package — may execute untrusted code* next `npx -y <pkg>@latest` fetches the newer one. Running that commandFix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packagesrc/plugin/launch.mjs:6 · npx invocation of non-well-known package — may execute untrusted code* `npx -y @ooples/token-optimizer-mcp@latest`. That keeps the deliberateFix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packagesrc/plugin/launch.mjs:17 · npx invocation of non-well-known package — may execute untrusted code* warm npx -y ...@latest -> ~6s (npx env setup + registry round-trip floor)Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
Process kill commanddist/proxy/server.js:218 · Process kill command — agent can terminate processes* and `Buffer.concat` over a stream nobody bounded is an out-of-memory kill with noFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
npx invocation of non-well-known packagedist/update/cli.js:11 · npx invocation of non-well-known package — may execute untrusted code* an npx cache entry, a Claude Code plugin, a source checkout, and an installFix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packagesrc/plugin/launch.mjs:13 · npx invocation of non-well-known package — may execute untrusted code* cold npx -y ...@latest -> ~18s to first response (blows the 30s MCPFix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packagedist/update/apply.js:77 · npx invocation of non-well-known package — may execute untrusted codereturn ('this copy is an npx cache entry, which is replaced rather than upgraded: ' +Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
Process kill commanddist/proxy/server.js:2132 · Process kill command — agent can terminate processes// only chance to record the tail of the window; a kill loses it, whichFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
npx invocation of non-well-known packagedist/server/daemon.js:9 · npx invocation of non-well-known package — may execute untrusted code* Performance: ~2-5ms per request (IPC overhead) vs ~1000-4000ms (npx spawn overhead)Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
Process kill commanddist/server/daemon.js:363 · Process kill command — agent can terminate processesconsole.error(`[DAEMON] Stop it first if you meant to restart: kill ${running}`);Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05

Cost to the agent

Instructions loaded
≈106 tok
1 file · 81 in bundle
Tool list per prompt
≈0 tok
0 tools · 0 per tool
Handshake
no reply
launch to tools/list, stdio
Undocumented tools
0
description under 20 chars · avg 0

Tokens estimated at four characters each. The tool list is re-read by the agent on every turn, so it is the number to compare between servers that do the same job: ranked across the index.

Known vulnerabilities

No advisories in OSV for @ooples/token-optimizer-mcp@7.11.0 as of 24 min ago. This is the package itself, not its dependencies.

What would make this Verified

10 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%
  1. Unclaimed publisher≈ +8 on the totalClaim ooples: 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. Access to sensitive file paths≈ +4 on the totalStop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
  4. Privileged sudo command≈ +4 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
  5. Credential pass-through configuration≈ +4 on the totalStop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
  6. Process kill command≈ +2 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
Publish a new version to npm, then queue it here. Not your skill? Claim the handle to get verdict changes by email.Claim ooplesPropose a patch to the maintainer