owlmem@0.9.3
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 4d ago · by lpm-firewall-aiThe package’s credential and shell-execution indicators have legitimate, gated uses. However, auto-wake silently reports AI-app usage totals to OwlMem, and that runtime telemetry is not disclosed in the README, so this warrants a warning.
Static reason
High-risk behavior combination matched malicious policy.
Trigger
Starts after the user enables auto-wake; the first usage report is scheduled 30 seconds after startup, then every 10 minutes.
Impact
OwlMem receives AI-app usage and quota information without disclosure in the README; no transcript content is included in the usage report.
Mechanism
Reads local Claude Code and Codex session records, aggregates token counts by day and app, and posts usage totals and status to OwlMem. Claude quota lookup is separately gated by `read_quota` and sends the sign-in token to Anthropic.
AI rationale
The package’s credential and shell-execution indicators have legitimate, gated uses. However, auto-wake silently reports AI-app usage totals to OwlMem, and that runtime telemetry is not disclosed in the README, so this warrants a warning.
Evidence
owlmem.mjsREADME.md~/.claude/projects/**/*.jsonl~/.codex/sessions/**/*.jsonl~/.config/owlmem/wake/usage.json
Network endpoints2
owlmem.com/v1/wake/usage (base URL can be overridden with OWLMEM_URL)api.anthropic.com/api/oauth/usage
Decision evidence
public snapshotAI called this Suspicious at 91.0% confidence as Unknown with medium false-positive risk.
Evidence for warning
- When auto-wake runs, the package periodically sends daily token-usage totals and status to OwlMem.
- The README describes auto-wake but does not disclose this usage reporting.
- The report aggregates token counts by day and app; the source does not send transcript text in this usage payload.
- Claude credentials are sent to Anthropic’s own usage endpoint only when the paired app’s `read_quota` setting is enabled.
Evidence against
- The manifest has no lifecycle scripts; the package is a CLI.
- Auto-wake is an explicit setup feature, and the package’s core vault requests go to OwlMem.
- The Claude credential flow is gated by the owner’s `read_quota` setting and targets `api.anthropic.com`.
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 = "0.9.3";
L45: const BASE = (process.env.OWLMEM_URL || "https://owlmem.com").replace(/\/$/, "");
L46: const DIR = process.env.OWLMEM_HOME || join(process.env.XDG_CONFIG_HOME || joi
Critical
Credential Exfiltration
Source appears to send environment or credential material to an external endpoint.
owlmem.mjsView on unpkg · L1414Trigger-reachable chain: manifest.bin -> owlmem.mjs
L14: // owlmem vault ls
L15: // owlmem vault get <name> [field] prints the item (or one field) to stdout
L16: // owlmem vault put <name> [--desc text] < value-or-json
...
L37: import { readFileSync, writeFileSync, mkdirSync, existsSync, chmodSync, copyFileSync, rmSync, statSync, realpathSync, openSync, closeSync, appendFileSync, renameSync, readdirSync, ...
L38: import { gunzipSync } from "node:zlib";
L39: import { homedir, hostname, userInfo } from "node:os";
...
L41: import { fileURLToPath } from "node:url";
L42: import { spawn, spawnSync } from "node:child_process";
L43:
L44: export const VERSION = "0.9.3";
L45: const BASE = (process.env.OWLMEM_URL || "https://owlmem.com").replace(/\/$/, "");
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@0.9.3.
See all versions of owlmem for other recorded verdicts.
- Review the evidence and your use of owlmem@0.9.3 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.