owlmem@0.9.7
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 3d ago · by lpm-firewall-aiThe inspected credential flow uses Claude Code’s token only for Anthropic’s usage endpoint, and only when the owner’s current OwlMem settings enable quota reading. The package is a CLI with no lifecycle scripts, and its entry point does not run the command handler when imported.
Static reason
High-risk behavior combination matched malicious policy.
Trigger
CLI commands; quota lookup requires auto-wake and a current owner-configured read_quota setting.
Impact
No unauthorized credential transfer or unrelated collection was found in the reviewed flows.
Mechanism
OwlMem API requests use the configured OwlMem service; Claude’s OAuth token is used for an Anthropic quota request.
AI rationale
The inspected credential flow uses Claude Code’s token only for Anthropic’s usage endpoint, and only when the owner’s current OwlMem settings enable quota reading. The package is a CLI with no lifecycle scripts, and its entry point does not run the command handler when imported.
Evidence
~/.claude/.credentials.json
Network endpoints3
owlmem.commcp.owlmem.comapi.anthropic.com
Decision evidence
public snapshotAI called this Clean at 92.0% confidence as Benign with low false-positive risk.
Evidence for AI clean decision
- The manifest defines a CLI entry point and no lifecycle scripts.
- The entry point calls main only when run as a command, not when imported.
- Claude credentials are read to obtain the Claude Code OAuth access token; quota lookup sends it to Anthropic’s own usage endpoint.
- The Anthropic quota lookup runs only after a current OwlMem poll reports read_quota for Claude Code.
- The README describes the package’s purpose as connecting AI apps to OwlMem and explains that sign-in requires the user to click Allow.
Evidence against
- The CLI has filesystem, network, and process execution capabilities, including a vault tool that can run a user-provided shell command.
- Auto-wake can run tasks in the background after the user enables and pairs it.
Behavioral surface
ChildProcessCryptoEnvironmentVarsFilesystemNetworkShell
HighEntropyStringsUrlStrings
Source & flagged code
10 flaggedowlmem.mjsView file
14// owlmem vault ls
L15: // owlmem vault get <name> [field] prints the item (or one field) to stdout
L16: // owlmem vault put <name> [--desc text] < value-or-json
...
L37: import { readFileSync, writeFileSync, mkdirSync, existsSync, chmodSync, copyFileSync, rmSync, statSync, realpathSync, openSync, closeSync, appendFileSync, renameSync, readdirSync, ...
L38: import { gunzipSync } from "node:zlib";
L39: import { homedir, hostname, userInfo } from "node:os";
...
L41: import { fileURLToPath } from "node:url";
L42: import { spawn, spawnSync } from "node:child_process";
L43:
L44: export const VERSION = "0.9.7";
L45: const BASE = (process.env.OWLMEM_URL || "https://owlmem.com").replace(/\/$/, "");
L46: const DIR = process.env.OWLMEM_HOME || join(process.env.XDG_CONFIG_HOME || joi
Critical
Credential Exfiltration
Source appears to send environment or credential material to an external endpoint.
owlmem.mjsView on unpkg · L1414Trigger-reachable chain: manifest.bin -> owlmem.mjs
L14: // owlmem vault ls
L15: // owlmem vault get <name> [field] prints the item (or one field) to stdout
L16: // owlmem vault put <name> [--desc text] < value-or-json
...
L37: import { readFileSync, writeFileSync, mkdirSync, existsSync, chmodSync, copyFileSync, rmSync, statSync, realpathSync, openSync, closeSync, appendFileSync, renameSync, readdirSync, ...
L38: import { gunzipSync } from "node:zlib";
L39: import { homedir, hostname, userInfo } from "node:os";
...
L41: import { fileURLToPath } from "node:url";
L42: import { spawn, spawnSync } from "node:child_process";
L43:
L44: export const VERSION = "0.9.7";
L45: const BASE = (process.env.OWLMEM_URL || "https://owlmem.com").replace(/\/$/, "");
L46: const DIR = process.env.OW
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@0.9.7.
See all versions of owlmem for other recorded verdicts.
- Review the evidence and your use of owlmem@0.9.7 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.