registry / @flydocs/cli / 1.2.14
@flydocs/cli@1.2.14
FlyDocs AI CLI — install, setup, and manage FlyDocs projects
AI Security Review
scanned 3mo ago · by lpm-firewall-aiThe package is not concretely malicious, but it installs first-party agent hooks that automatically collect and upload AI usage metadata after workspace consent. This warrants a warning under the agent-extension lifecycle policy.
Static reason
One or more suspicious static signals were detected.
Trigger
User runs FlyDocs setup, then a configured Claude session reaches Stop.
Impact
AI-session metadata and token counters may leave the machine when the workspace enables collection.
Mechanism
project agent hooks invoke consent-gated usage reporting
AI rationale
The package is not concretely malicious, but it installs first-party agent hooks that automatically collect and upload AI usage metadata after workspace consent. This warrants a warning under the agent-extension lifecycle policy.
Evidence
package.jsondist/cli.jstemplate/.claude/settings.jsontemplate/.claude/hooks/usage-capture.py.claude/settings.json.claude/hooks/usage-capture.py~/.claude/projects~/.codex~/.cursor
Network endpoints2
app.flydocs.ai/api/relay/usage/sessionsapi.anthropic.com/api/oauth/usage
Decision evidence
public snapshotAI called this Suspicious at 90.0% confidence as Dangerous Capability with low false-positive risk.
Evidence for warning
- template/.claude/settings.json installs Stop hooks after explicit CLI setup.
- usage-capture.py launches `flydocs usage report` on each agent Stop event.
- dist/cli.js reads local Claude/Codex/Cursor session records and can POST usage data.
- dist/cli.js reads Claude OAuth credentials solely to query Anthropic usage status.
Evidence against
- package.json has no preinstall, install, or postinstall hook.
- Usage upload is gated by workspace consent, API credentials, and a machine opt-out.
- Wire records contain usage counters/metadata; inspected serializer does not include transcript text.
- Agent files are project-local FlyDocs templates, not install-time broad host mutation.
Behavioral surface
ChildProcessCryptoDynamicRequireEnvironmentVarsFilesystemNetworkShell
HighEntropyStringsUrlStrings
Source & flagged code
9 flaggeddist/cli.jsView file
189});
L190: import { execSync } from "child_process";
L191: import { access as access2, constants } from "fs/promises";
High
8355const { promisify: promisify6 } = await import("util");
L8356: const execAsync = promisify6(execFile6);
L8357: await execAsync("python3", [scriptPath, "pull"], {
High
189});
L190: import { execSync } from "child_process";
L191: import { access as access2, constants } from "fs/promises";
...
L248: const { readFile: readFile31, writeFile: writeFile25, mkdir: mkdir23 } = await import("fs/promises");
L249: const cacheDir = join46(process.env.HOME ?? "~", ".flydocs");
L250: const cachePath2 = join46(cacheDir, "version-check.json");
L251: try {
L252: const cache = JSON.parse(await readFile31(cachePath2, "utf-8"));
L253: const age = Date.now() - new Date(cache.checkedAt ?? 0).getTime();
...
L413: console.log(
L414: ` ${dim("Docs and updates:")} ${pc2.cyan("https://www.flydocs.ai")}`
L415: );
High
Credential Exfiltration
Source combines credential-like environment material and outbound requests; review data flow before blocking.
Findings
7 High5 Medium5 Low
HighChild Processdist/cli.js
HighShelldist/cli.js
HighCredential Exfiltrationdist/cli.js
HighSandbox Evasion Gated Capabilitydist/cli.js
HighTrigger Reachable Credential Exfiltrationdist/cli.js
HighRuntime Package Installdist/cli.js
HighPayload In Excluded Dirtemplate/.claude/hooks/repo_context.py
MediumDynamic Requiredist/cli.js
MediumNetwork
MediumEnvironment Vars
MediumShips Build Helpertemplate/.claude/hooks/repo_context.py
MediumStructural Risk Force Deep Review
LowScripts Present
LowFilesystem
Affected versions and remediation
This report applies to @flydocs/cli@1.2.14.
See all versions of @flydocs/cli for other recorded verdicts.
- Review the evidence and your use of @flydocs/cli@1.2.14 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.