owlmem@0.9.14
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 7h ago · by lpm-firewall-aiAuto-wake periodically reads local AI transcript metadata and sends aggregated token usage to OwlMem, but the README does not disclose this runtime telemetry. Auto-wake is opt-in, and the Claude OAuth token flow is limited to Anthropic’s own quota endpoint, so the evidence supports a warning rather than a block.
Static reason
High-risk behavior combination matched malicious policy.
Trigger
Runs after a user enables and starts auto-wake.
Impact
Reveals per-app usage and activity patterns to the OwlMem service and agents in the space.
Mechanism
Reads local Claude Code and Codex transcript/session metadata, aggregates token counts and quota status, then posts the report to OwlMem.
AI rationale
Auto-wake periodically reads local AI transcript metadata and sends aggregated token usage to OwlMem, but the README does not disclose this runtime telemetry. Auto-wake is opt-in, and the Claude OAuth token flow is limited to Anthropic’s own quota endpoint, so the evidence supports a warning rather than a block.
Evidence
owlmem.mjsREADME.md~/.claude/projects/**/*.jsonl~/.codex/sessions/**/*.jsonlusage.json
Network endpoints2
owlmem.com/v1/wake/usageapi.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
- Auto-wake reads Claude Code and Codex transcripts or sessions, aggregates token usage by app and day, and reports it to OwlMem so agents can see activity.
- The usage report is posted to the OwlMem wake API with daily usage and status data; auto-wake sends it shortly after startup and then periodically.
- The README explains how to enable auto-wake but does not disclose the transcript-based usage reporting.
Evidence against
- Auto-wake is an explicit feature enabled with `--wake` or `owlmem wake enable`; `--yes` alone leaves it off.
- Claude OAuth credentials are sent to Anthropic’s own quota endpoint only when the owner enables Claude quota reading; OwlMem receives quota percentages and reset times.
- The package manifest contains no lifecycle scripts.
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 = "0.9.14";
L47: const BASE = (process.env.OWLMEM_URL || "https://owlmem.com").replace(/\/$/, "");
L48: const DIR = process.env.OWLMEM_HOME || join(process.env.XDG_CONFIG_HOME || jo
Critical
Credential Exfiltration
Source appears to send environment or credential material to an external endpoint.
owlmem.mjsView on unpkg · L1616Trigger-reachable chain: manifest.bin -> owlmem.mjs
L16: // owlmem vault ls
L17: // owlmem vault get <name> [field] prints the item (or one field) to stdout
L18: // owlmem vault put <name> [--desc text] < value-or-json
...
L39: import { readFileSync, writeFileSync, mkdirSync, existsSync, chmodSync, copyFileSync, rmSync, statSync, realpathSync, openSync, closeSync, appendFileSync, renameSync, readdirSync, ...
L40: import { gunzipSync } from "node:zlib";
L41: import { homedir, hostname, userInfo } from "node:os";
...
L43: import { fileURLToPath } from "node:url";
L44: import { spawn, spawnSync } from "node:child_process";
L45:
L46: export const VERSION = "0.9.14";
L47: const BASE = (process.env.OWLMEM_URL || "https://owlmem.com").replace(/\/$/, "");
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@0.9.14.
See all versions of owlmem for other recorded verdicts.
- Review the evidence and your use of owlmem@0.9.14 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.