registry / iloso / 0.4.0

iloso@0.4.0

A local-first social publishing CLI and MCP server for AI agents.

AI Security Review

scanned 1d ago · by lpm-firewall-ai

LPM treats this as warn-only first-party agent extension lifecycle risk. No install-time execution or concrete malicious chain is established. An explicit CLI skill-install command delegates agent-skill installation to an externally resolved `npx` command; normal runtime activity manages local X publishing data and X OAuth/API calls.

Static reason

No blocking static signals were detected.

Trigger

User runs `ilo skill install`; other network and storage actions require explicit CLI or MCP use.

Impact

Potential agent-extension lifecycle mutation from the explicit command; no unconsented install-time control-surface change was found.

Mechanism

Explicit external agent-skill installer plus local X OAuth/publishing client.

Rationale

Source inspection found no concrete malicious behavior, but the explicit agent-skill installer executes an externally resolved command and may mutate agent setup. Per firewall policy, this warrants a warning rather than a block.

Evidence

package.jsondist/cli.jsdist/src-CLoHebRH.jsdist/mcp.jsskills/ilo/SKILL.mdREADME.md.config/ilo/config.json.config/ilo/ilo.sqlite

Network endpoints2

api.x.com/2x.com/i/oauth2/authorize

iloso@0.4.0

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Source & flagged code