owlmem@0.9.10
OwlMem command line: connect your AI apps to OwlMem in one command (npx -y owlmem install), an end-to-end encrypted vault for your AI agents, and a local MCP server.
AI Security Review
scanned 2d ago · by lpm-firewall-aiThe flagged credential flow is for Claude quota reporting: it is gated by an owner-enabled setting and sends the Claude token to Anthropic’s own usage endpoint. The package’s executable dispatches explicit CLI commands, and no install-time script or unrelated credential destination was found.
Static reason
High-risk behavior combination matched malicious policy.
Trigger
The CLI runs when invoked; quota reporting occurs in configured auto-wake usage checks, and Claude quota access is gated by an owner-enabled setting.
Impact
The inspected credential flow supports quota reporting for the package’s stated purpose; no credential exfiltration or unrelated automatic action was found.
Mechanism
The CLI sends its OwlMem token to the configured OwlMem service. When enabled, it reads Claude Code’s credential and sends that token to Anthropic’s usage endpoint, then reports quota status to OwlMem.
AI rationale
The flagged credential flow is for Claude quota reporting: it is gated by an owner-enabled setting and sends the Claude token to Anthropic’s own usage endpoint. The package’s executable dispatches explicit CLI commands, and no install-time script or unrelated credential destination was found.
Evidence
owlmem.mjspackage.json
Network endpoints3
owlmem.commcp.owlmem.com/mcpapi.anthropic.com/api/oauth/usage
Decision evidence
public snapshotAI called this Clean at 94.0% confidence as Benign with low false-positive risk.
Evidence for AI clean decision
- The manifest identifies this as OwlMem’s CLI and declares owlmem.mjs as its executable entrypoint, with no lifecycle scripts.
- Claude Code credentials are read for quota reporting, which the source says requires the owner to enable “Read Claude quota” in the console.
- The Claude token is sent to Anthropic’s own usage endpoint; the source says OwlMem receives only quota percentages and reset times.
- OwlMem API requests send the configured OwlMem token to the OwlMem service.
- The CLI dispatches work only when invoked as the executable, then routes commands through explicit command handlers.
Evidence against
Behavioral surface
ChildProcessCryptoEnvironmentVarsFilesystemNetworkShell
HighEntropyStringsUrlStrings
Source & flagged code
10 flaggedowlmem.mjsView file
16// owlmem vault ls
L17: // owlmem vault get <name> [field] prints the item (or one field) to stdout
L18: // owlmem vault put <name> [--desc text] < value-or-json
...
L39: import { readFileSync, writeFileSync, mkdirSync, existsSync, chmodSync, copyFileSync, rmSync, statSync, realpathSync, openSync, closeSync, appendFileSync, renameSync, readdirSync, ...
L40: import { gunzipSync } from "node:zlib";
L41: import { homedir, hostname, userInfo } from "node:os";
...
L43: import { fileURLToPath } from "node:url";
L44: import { spawn, spawnSync } from "node:child_process";
L45:
L46: export const VERSION = "0.9.10";
L47: const BASE = (process.env.OWLMEM_URL || "https://owlmem.com").replace(/\/$/, "");
L48: const DIR = process.env.OWLMEM_HOME || join(process.env.XDG_CONFIG_HOME || jo
Critical
Credential Exfiltration
Source appears to send environment or credential material to an external endpoint.
owlmem.mjsView on unpkg · L1616Trigger-reachable chain: manifest.bin -> owlmem.mjs
L16: // owlmem vault ls
L17: // owlmem vault get <name> [field] prints the item (or one field) to stdout
L18: // owlmem vault put <name> [--desc text] < value-or-json
...
L39: import { readFileSync, writeFileSync, mkdirSync, existsSync, chmodSync, copyFileSync, rmSync, statSync, realpathSync, openSync, closeSync, appendFileSync, renameSync, readdirSync, ...
L40: import { gunzipSync } from "node:zlib";
L41: import { homedir, hostname, userInfo } from "node:os";
...
L43: import { fileURLToPath } from "node:url";
L44: import { spawn, spawnSync } from "node:child_process";
L45:
L46: export const VERSION = "0.9.10";
L47: const BASE = (process.env.OWLMEM_URL || "https://owlmem.com").replace(/\/$/, "");
L48: const DIR = process.env.O
Findings
2 Critical8 High3 Medium3 Low
CriticalCredential Exfiltrationowlmem.mjs
CriticalTrigger Reachable Dangerous Capabilityowlmem.mjs
HighChild Processowlmem.mjs
HighShellowlmem.mjs
HighSame File Env Network Executionowlmem.mjs
HighCommand Output Exfiltrationowlmem.mjs
HighSandbox Evasion Gated Capabilityowlmem.mjs
HighTrigger Reachable Credential Exfiltrationowlmem.mjs
HighTrigger Reachable Command Output Exfiltrationowlmem.mjs
HighSemantic Analysis Limitedowlmem.mjs
MediumNetwork
MediumEnvironment Vars
MediumStructural Risk Force Deep Review
Affected versions and remediation
This report applies to owlmem@0.9.10.
See all versions of owlmem for other recorded verdicts.
- Review the evidence and your use of owlmem@0.9.10 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.