# securemyagent.ai

> SMA is the security layer for AI agents. An agent registers itself, runs a free Checkup, gets a signed 0-100 score, and keeps itself current with a heartbeat.

Start here: https://securemyagent.ai/skill.md. Tell your agent: `Read https://securemyagent.ai/skill.md and follow it`.

## Onboarding

- [skill.md](https://securemyagent.ai/skill.md): register, run a Checkup, claim, heartbeat
- [heartbeat.md](https://securemyagent.ai/heartbeat.md): what to run every 4-24 hours
- [openapi.json](https://securemyagent.ai/openapi.json): API spec
- [llms-full.txt](https://securemyagent.ai/llms-full.txt): full text of every page below

## Docs

- [Quickstart](https://securemyagent.ai/docs/quickstart.md): Register an agent, run a Checkup, show the score and claim URL, add the heartbeat. About 60 seconds.
- [API reference](https://securemyagent.ai/docs/api.md): Every v1 endpoint, its auth, and the M1 request and response shapes. Machine-readable copy at /openapi.json.
- [Scoring formula](https://securemyagent.ai/docs/scoring.md): How the Agent Security Score (0-100) is computed. Public, versioned, reproducible from a report.
- [How SMA protects itself](https://securemyagent.ai/docs/self-protection.md): The rules SMA holds itself to. Secrets, keys, supply chain, crawlers, tenants, abuse.

## Runtime guides

- [Secure an OpenClaw agent](https://securemyagent.ai/articles/secure-openclaw.md): Register OpenClaw with SMA, run the Checkup, lock the gateway, and add the heartbeat.
- [Secure a Claude Code agent](https://securemyagent.ai/articles/secure-claude-code.md): Register Claude Code with SMA, tighten permissions, vet MCP servers, and schedule the heartbeat.
- [Secure a Hermes agent](https://securemyagent.ai/articles/secure-hermes.md): Register a Hermes agent with SMA, run the Checkup from its terminal tool, and schedule the heartbeat.
- [Secure a Codex agent](https://securemyagent.ai/articles/secure-codex.md): Register Codex CLI with SMA, set the sandbox and approval mode, vet MCP servers in config.toml, and add the heartbeat.
- [Secure a Cursor agent](https://securemyagent.ai/articles/secure-cursor.md): Register Cursor with SMA, audit mcp.json, turn off blind auto-run, and add the heartbeat.
- [Secure a custom MCP client](https://securemyagent.ai/articles/secure-custom-mcp-client.md): Onboard any MCP client with SMA over HTTP, and the client-side rules that stop tool poisoning.
- [Secure an n8n agent workflow](https://securemyagent.ai/articles/secure-n8n.md): Register an n8n workflow as an SMA agent, lock down the n8n host, and run the heartbeat from a Schedule trigger.

## Fix guides, one per Checkup check

- [Fix: your gateway is on 0.0.0.0](https://securemyagent.ai/articles/fix-network-exposure.md): Find agent services listening on all interfaces and bind them to localhost.
- [Fix: your runtime or MCP server has a known CVE](https://securemyagent.ai/articles/fix-version-cves.md): Find vulnerable versions of your agent runtime and MCP packages, and update them.
- [Fix: plaintext API keys in config files](https://securemyagent.ai/articles/fix-secrets-at-rest.md): Find keys in configs, .env files and memory files. Move them out. Rotate them.
- [Fix: a skill or MCP server has a bad verdict](https://securemyagent.ai/articles/fix-installed-components.md): Remove skills and MCP servers that are malicious or unknown, and vet new ones before install.
- [Fix: shell, network and file-write with no approval](https://securemyagent.ai/articles/fix-tool-scope.md): Cut tool permissions to what the agent needs and require approval for risky calls.
- [Fix: no egress allowlist](https://securemyagent.ai/articles/fix-egress.md): Limit which hosts the agent can reach so a hijacked agent cannot send your data out.
- [Fix: no stable identity and no audit log](https://securemyagent.ai/articles/fix-identity-audit.md): Give the agent a stable DID and log every tool call.
- [Fix: untrusted content in long-term memory](https://securemyagent.ai/articles/fix-memory-context.md): Stop web and tool content from becoming trusted memory. Record where each memory came from.

## Threat explainers

- [ClawHavoc - how malicious skills reached OpenClaw agents](https://securemyagent.ai/articles/clawhavoc-skill-malware.md): 341 of 2,857 audited ClawHub skills were malicious. What they did and how to check yours.
- [Exposed gateways and CVE-2026-25253](https://securemyagent.ai/articles/exposed-gateways-cve-2026-25253.md): OpenClaw gateways on port 18789 bound to 0.0.0.0 with auth off. A one-click RCE with CVSS 8.8.
- [MCP tool poisoning and rug pulls](https://securemyagent.ai/articles/mcp-tool-poisoning-rug-pulls.md): Hidden instructions in tool descriptions, and servers that change after you approved them.
- [The Moltbook token leak](https://securemyagent.ai/articles/moltbook-token-leak.md): A Supabase key in client JavaScript with row-level security off exposed about 1.5M agent tokens.
- [Prompt-injection exfiltration](https://securemyagent.ai/articles/prompt-injection-exfiltration.md): How a web page, form or ticket makes an agent send your data to an attacker, and how to cut the path.