cli@1.6.3
CAUTION· install onlyPublisher claims: "Connects Claude Code, Codex, pi and other coding agents to Phantom AI, an OpenAI-compatible API, and manages your keys, credit and spending caps from the terminal." Observed in sandbox: no network, touches nothing outside the project, writes no memory.
Coverage
Bundled scripts ran with no arguments; no agent task scenario for this bundle.- Tested
- 1 of 1 bundled script 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 1.6.0
vet-skills diff cli 1.6.0 1.6.3In the sandbox, this version reached for nothing 1.6.0 did not. Compared from the two recorded runs, not from source.
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 2026-11-18 and scored it 67/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Collapsed <details> block carrying instructions; Credential pass-through configuration; Credential pass-through configuration. Summary: no network, touches nothing outside the project, writes no memory.
Behavior trace
Honeypot run · 0 s · 2 scenarios · 0 deviationsStatic findings
// server would kill the recovery file, so a sign-in that is the recovery credential is only forgotten here.Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.<summary><strong>Development and contributing</strong></summary> The CLI lives in `cli/` in the Phantom repo, which uses pnpm. Install once at the repo root, thFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.export function proxyToken() {Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.`curl -fsSL https://phantom.codes/install | sh, which brings its own.\n`);Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.const token = proxyToken();Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.* it started: each request must carry the run's token ({@link proxyToken}),Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.import { proxyToken, startProxy } from "./proxy.mjs";Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.* @param opts.token The run's secret ({@link proxyToken}). A request without it, as `Authorization: Bearer` or `x-api-key`, gets 401.Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.`curl -fsSL https://phantom.codes/install | sh, which brings its own.\n`);Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.// runs the helper outside this shell, so a relative or npx path won't do.Fix: Pin versions and sources; vendor anything fetched at install time.Cost to the agent
Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.
Known vulnerabilities
No advisories in OSV for @phantomcodes/cli@1.6.3 as of 7 min ago. This is the package itself, not its dependencies.
What would make this Verified
13 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%- Pipe-to-shell pattern≈ +9 on the totalBuild commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
- Unclaimed publisher≈ +8 on the totalClaim phantomcodes: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
- 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.
- 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.
- Process kill command≈ +2 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
- Remote installer piped to shell≈ +2 on the totalBuild commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.