osborn@0.9.224
Voice AI coding assistant - local agent that connects to Osborn frontend
AI Security Review
scanned 27d ago · by lpm-firewall-aiThe package performs concrete Claude control-surface mutation when its command is run, so it warrants a warning. The inspected metadata contains no registry install hook and no confirmed credential exfiltration.
Static reason
High-risk behavior combination matched malicious policy.; previous stored version diff introduced dangerous source
Trigger
A user runs the osborn command.
Impact
This can alter an existing AI-agent configuration and redirect its stored state.
Mechanism
The command launches the agent, whose imported Claude module updates settings at module load; the launcher can create a Claude-directory symlink on matching hosts.
AI rationale
The package performs concrete Claude control-surface mutation when its command is run, so it warrants a warning. The inspected metadata contains no registry install hook and no confirmed credential exfiltration.
Evidence
bin/cli.jsdist/claude-llm.jsdist/index.js~/.claude/settings.json~/.claude/workspace/.claude
Decision evidence
public snapshotAI called this Suspicious at 90.0% confidence as Dangerous Capability with low false-positive risk.
Evidence for warning
- The package exposes an osborn command that launches its agent.
- Launching the command runs the compiled agent entry point, which imports the Claude integration.
- The imported Claude module automatically changes the user's Claude settings at module load.
- The command also conditionally repoints the user's Claude directory to a workspace location.
Evidence against
- package.json has no install or postinstall lifecycle hook; its only publish hook builds the package.
- The inspected OAuth handling persists a token locally for the Claude SDK, with no inspected exfiltration sink.
Behavioral surface
ChildProcessCryptoDynamicRequireEnvironmentVarsFilesystemNetworkShell
HighEntropyStringsUrlStrings
Source & flagged code
9 flagged · loading sourcebin/recall.jsView file
5L6: import { spawn } from 'child_process'
L7: import { fileURLToPath } from 'url'
High
dist/jsonl-search.jsView file
34// @vscode/ripgrep bundles the rg binary — this is our own dependency
L35: const { rgPath } = require('@vscode/ripgrep');
L36: if (rgPath && existsSync(rgPath))
Medium
Dynamic Require
Package source references dynamic require/import behavior.
dist/jsonl-search.jsView on unpkg · L34dist/index.jsView file
2529const which = await new Promise((res) => {
L2530: const p = spawn('which', ['code-server']);
L2531: let out = '';
...
L2538: await new Promise((res, rej) => {
L2539: const install = spawn('sh', ['-c', 'curl -fsSL https://code-server.dev/install.sh | sh'], {
L2540: env: { ...process.env },
L2541: stdio: ['ignore', 'pipe', 'pipe'],
Findings
5 Critical3 High5 Medium5 Low
CriticalSame File Env Network Executiondist/index.js
CriticalCredential Exfiltrationdist/index.js
CriticalCommand Output Exfiltrationdist/index.js
CriticalDownload Executedist/index.js
CriticalPrevious Version Dangerous Deltadist/index.js
HighChild Processbin/recall.js
HighShell
HighSemantic Analysis Limiteddist/session-access.js#virtual:normalized:round1
MediumDynamic Requiredist/jsonl-search.js
MediumNetwork
MediumEnvironment Vars
MediumShips Build Helperdeploy.sh
MediumStructural Risk Force Deep Review
Affected versions and remediation
This report applies to osborn@0.9.224.
See all versions of osborn for other recorded verdicts.
- Review the evidence and your use of osborn@0.9.224 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.