Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/zhhjluka/hm-arch/openclawgit clone --depth 1 https://github.com/zhhjluka/hm-archWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/zhhjluka/hm-arch/openclaw)<a href="https://agentmods.dev/agents/zhhjluka/hm-arch/openclaw"><img src="https://agentmods.dev/badge/agents/zhhjluka/hm-arch/openclaw.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00000 | $0.02159 |
| Opus 5 | $0.00000 | $0.01079 |
| Sonnet 5 | $0.00000 | $0.00432 |
| Haiku 4.5 | $0.00000 | $0.00216 |
Grade A, and why
openclaw scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 267 lines — stays where its author put it; the contents beside it link to each section on GitHub.
OpenClaw setup (HM-Arch)
Install HM-Arch as OpenClaw's native memory provider. OpenClaw integration uses
the @hm-arch/openclaw-plugin TypeScript memory plugin backed by a persistent
Python sidecar that exposes HM-Arch recall, capture, forget, and consolidation
without per-query process startup.
All management commands are offline and do not require LLM/API keys.
Current availability
| Path | Status | Notes |
|---|---|---|
hm-arch install openclaw |
Available | Registers config, plugin package, sidecar command, and SQLite database. |
hm-arch status openclaw / doctor openclaw |
Available | Inspect slot, sidecar command, storage, and plugin manifest. |
hm-arch uninstall openclaw |
Available | Removes HM-Arch-owned config and plugin files; preserves the database. |
npx @hm-arch/installer install openclaw |
Available | Same behavior via the npm launcher (hm-arch-install). |
hm-arch install openclaw is registered in the CLI alongside codex, claude-code,
and hermes. On success it writes openclaw.json plugin settings and copies the
canonical @hm-arch/openclaw-plugin package under the OpenClaw extensions directory.
Memory modes
OpenClaw can run with several memory configurations. HM-Arch documentation and benchmarks treat these as distinct modes:
| Mode | Description |
|---|---|
| No memory | OpenClaw runs without an active memory provider. Baseline for benchmarks. |
| Native memory | OpenClaw's built-in memory slot uses another provider already configured in OpenClaw. |
| HM-Arch | The HM-Arch memory plugin owns the OpenClaw memory slot and delegates durable storage to the Python sidecar. |
| Mem0 | External Mem0 provider configured in OpenClaw instead of HM-Arch. |
| OpenViking | External OpenViking provider configured in OpenClaw instead of HM-Arch. |
HM-Arch install commands detect an occupied memory slot and refuse to overwrite another provider silently. Switch providers explicitly in OpenClaw configuration when you intend to replace Mem0, OpenViking, or another native backend.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago First seen · 267 lines · 0 tokens per session scan A 3ac0e09baeff
openclaw is an agent published in the GitHub repository zhhjluka/hm-arch (18 stars, last pushed 2mo ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 2,159 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
history-management
The runtime keeps conversation history for each agent session and sends a provider-facing working history to the model. Two complementary limits operate on different representations.
design-reviewer
Design-review sub-agent. Reviews design docs across the four dimensions of architecture, interface, performance, and security, covering MemOS's multi-memory / multi-storage backend constraints.
integration-tester
MemOS integration-testing sub-agent. Authors and executes pytest cases under tests/ based on the task's requirements and design, and emits real test reports.
explorer
Read-only code exploration sub-agent. Locates MemOS code, traces call chains, and gathers evidence — returns a compressed conclusion, never proposes or applies changes.
Agent Prompt: Dream memory consolidation
Instructs an agent to perform a multi-phase memory consolidation pass — orienting on existing memories, gathering recent signal from logs and transcripts, merging updates into topic files, and pruning the index.
lessons-learned
Hard-won knowledge from building this codebase. When you make a mistake or discover a non-obvious behavior, add it here.