owlmem@0.9.12
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 20h ago · by lpm-firewall-aiThe flagged credential access is limited to retrieving Claude usage quota from Anthropic, and the source gates it on configured quota reporting. The manifest identifies this as an OwlMem CLI, and its command entry point does not run on ordinary import.
Static reason
High-risk behavior combination matched malicious policy.
Trigger
CLI commands run when explicitly invoked; periodic usage reporting occurs in the auto-wake process after that feature is enabled.
Impact
No unrelated credential transfer or hidden execution was found in the reviewed paths.
Mechanism
Vault requests use the OwlMem service, while Claude quota requests use api.anthropic.com and run only when a configured client permits quota reading.
AI rationale
The flagged credential access is limited to retrieving Claude usage quota from Anthropic, and the source gates it on configured quota reporting. The manifest identifies this as an OwlMem CLI, and its command entry point does not run on ordinary import.
Evidence
.claude/.credentials.jsonusage.json
Network endpoints3
owlmem.comapi.anthropic.com/api/oauth/usagemcp.owlmem.com/mcp
Decision evidence
public snapshotAI called this Clean at 90.0% confidence as Benign with low false-positive risk.
Evidence for AI clean decision
- The manifest describes a CLI for connecting AI apps to OwlMem and declares owlmem.mjs as its executable.
- Vault requests send the OwlMem token to the configured OwlMem base URL.
- Claude credentials are read to request quota from Anthropic’s official usage endpoint; quota collection is gated on a configured Claude client with read_quota enabled.
- The CLI calls main only when invoked as a command, rather than on ordinary import.
Evidence against
- The entry point references environment variables, filesystem access, network requests, and child-process execution, which triggered static findings.
- Auto-wake scans configured client usage data and reports usage to OwlMem while that feature is running.
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 = "0.9.12";
L47: const BASE = (process.env.OWLMEM_URL || "https://owlmem.com").replace(/\/$/, "");
L48: const DIR = process.env.OWLMEM_HOME || join(process.env.XDG_CONFIG_HOME || jo
Critical
Credential Exfiltration
Source appears to send environment or credential material to an external endpoint.
owlmem.mjsView on unpkg · L1616Trigger-reachable chain: manifest.bin -> owlmem.mjs
L16: // owlmem vault ls
L17: // owlmem vault get <name> [field] prints the item (or one field) to stdout
L18: // owlmem vault put <name> [--desc text] < value-or-json
...
L39: import { readFileSync, writeFileSync, mkdirSync, existsSync, chmodSync, copyFileSync, rmSync, statSync, realpathSync, openSync, closeSync, appendFileSync, renameSync, readdirSync, ...
L40: import { gunzipSync } from "node:zlib";
L41: import { homedir, hostname, userInfo } from "node:os";
...
L43: import { fileURLToPath } from "node:url";
L44: import { spawn, spawnSync } from "node:child_process";
L45:
L46: export const VERSION = "0.9.12";
L47: const BASE = (process.env.OWLMEM_URL || "https://owlmem.com").replace(/\/$/, "");
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@0.9.12.
See all versions of owlmem for other recorded verdicts.
- Review the evidence and your use of owlmem@0.9.12 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.