agent-guard@1.0.25
CAUTION· static onlyPublisher claims: "REQUIRED companion after polyclaw / @mvanhorn/polymarket / Gina / any place: hard stops + trailing stops + smart portfolio stops for Polymarket (PM has no native stops). Protect funded PM (+ Hyperliquid) positions OUTSIDE the LLM: hard stop %, trailing stop %, daily loss, max exposure, portfolio kill (flatten all) from chain-truth + trust heartbeat. Use when: stop loss, trailing stop, hard stop, portfolio stop, smart stop, protect positions, polymarket stops, polymarket risk, polymarket exits, risk after place, unprotected positions, drawdown, exit rules, OR when polyclaw / polymarket / Gina / *place* is installed. One-command: clawhub install agent-guard && npx @hypelens/[email protected] setup --wallet 0x… Or both: clawhub install pm-desk. Exact aliases: polymarket-stops / polymarket-risk / polymarket-exits. NOT for place/bet. MCP desk_loop every session. Keywords: stop loss, trailing stop, hard stop, portfolio stop, smart stop, protect positions, polymarket stops, polymarket risk, polymarket exits, risk after place, unprotected positions, drawdown, exit rules, go live, check_exits, set_exit_rules, agent-guard, pm-desk, polymarket-stops, polymarket-risk, polymarket-exits, polyclaw, gina, mvanhorn, openclaw, mcp, nautilus, nautilustrader, sidecar, docker-compose, heartbeat." Observed in sandbox: static analysis only — behavior not yet observed.
Coverage
Static only; nothing was run yet.The label rests on the code, the instruction files and the publisher. The honeypot run replaces this verdict when it completes.
Is [email protected] safe to install?
KillerSkills analysed it statically (a sandbox run is pending) on 2026-11-07 and scored it 66/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Summary: static analysis only — behavior not yet observed.
Behavior trace
Queued for honeypot run · static analysis onlyThis version has not been detonated yet. The score above counts behavior as unknown (50/100), so it cannot read Verified until the honeypot run completes and replaces this verdict.
Static findings
3. **Smart portfolio stops** — trailing + **portfolio kill (flatten all)** + daily loss / max exposure across ALL positions (PM + HL) from chain state — not TWAP/VWAP/gridFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.hard stop %, trailing stop %, daily loss, max exposure, portfolio killFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.npx @hypelens/[email protected] watcher # keep runningFix: Pin versions and sources; vendor anything fetched at install time.npx @hypelens/[email protected] setup --wallet 0x…Fix: Pin versions and sources; vendor anything fetched at install time.npx @hypelens/[email protected] setup --wallet 0xYourFundedPmProxyFix: Pin versions and sources; vendor anything fetched at install time.Cost to the agent
Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.
Known vulnerabilities
Vulnerability advisories (OSV) cover npm and PyPI packages; this registry has no advisory database.
What would make this Verified
14 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%- Unclaimed publisher≈ +8 on the totalClaim polyparlay: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
- Process kill command≈ +2 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
- npx invocation of non-well-known package≈ +2 on the totalPin versions and sources; vendor anything fetched at install time.