ESP-Claw is an AI agent framework that runs on Espressif IoT chips and lets people define device behavior through conversation. It handles local sensing, decision-making, and execution for connected devices, with support for event-driven actions, structured memory, and MCP communication.
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 skills add espressif/esp-claw --skill memory_opsgit clone --depth 1 https://github.com/espressif/esp-clawWrote 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/skills/espressif/esp-claw/memory_ops)<a href="https://agentmods.dev/skills/espressif/esp-claw/memory_ops"><img src="https://agentmods.dev/badge/skills/espressif/esp-claw/memory_ops/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/espressif/esp-claw/memory_ops"><img src="https://agentmods.dev/badge/skills/espressif/esp-claw/memory_ops.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 5.1 | $0.00024 | $0.01313 |
| Opus 5.5 | $0.00010 | $0.00525 |
| Sonnet 5.5 | $0.00005 | $0.00263 |
| Haiku 4.5 | $0.00002 | $0.00131 |
Grade A, and why
memory_ops 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 today.
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 — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Long-term Memory
Use on-device long-term memory tools to remember, recall, list, update, and forget structured memories.
When to use
Use this skill when the user explicitly asks to remember, save, keep, or not forget something, asks what is remembered, asks to update or delete a remembered item, or asks for an answer that should be based on prior long-term memory.
Hard Rules
- If the user clearly asks to remember, save, keep, or not forget something, call
memory_store. - Write
memory_store.contentas a concise normalized memory fact, not the user's raw quote. - Do not write raw user quotes into memory content.
- Do not exceed the maximum number of retrieval terms.
- Keep retrieval terms in the same language as the memory fact unless the memory itself is primarily in another language.
- If the user asks what you remember, asks to verify a remembered fact, or asks for a personalized answer based on prior memory, inspect the injected summary labels first and call
memory_recallwhen relevant labels are present. - Do not place natural-language questions into
summary_labels. memory_recallrequires exactsummary_labelschosen from the injected catalog.- Use
queryonly to narrow the search within the selected summary labels. - Use
memory_listwhen the user wants to inspect stored memories. - Use
memory_updateonly when one existing memory should be modified. - Use
memory_forgetonly when one existing memory should be removed. - Use
memory_recallplus exactmemory_idonly when you are already doing an explicit memory-inspection or memory-editing flow and need to inspect the recalled memory bodies yourself. - Do not use non-whitelisted or free-text values as summary labels.
- Do not read or write
memory_records.jsonl,memory_index.json,memory_digest.log, orMEMORY.mdto make decisions directly. - Summary labels are not the memory body. Use
memory_recallto obtain detailed stored content. - Do not call
memory_storefor ordinary self-introductions or casual preference statements unless the user explicitly asks to save them. Let automatic extraction handle durable facts after the reply silently. - Do not make the whole reply an operation log such as “I have remembered” or “deleted” unless the user explicitly asked only for a memory operation.
- Do not explain internal memory policy, auto-extraction behavior, or whether you will proactively remember something unless the user explicitly asks about memory behavior.
- Do not ask whether the user wants you to remember ordinary profile or preference statements when automatic extraction can handle them. Do not offer memory-save help unless the user explicitly asks about memory management.
- Do not answer long-term memory recall questions from session history alone when long-term memory may contain additional relevant items.
- Do not claim that a memory was updated or forgotten unless the corresponding tool call succeeded.
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.
- today Changed · -1 lines c04d6bf07796
- 1mo ago First seen · 72 lines · 24 tokens per session scan A cdc2da05aa8b
memory_ops is a skill published in the GitHub repository espressif/esp-claw (2,260 stars, last pushed yesterday), licensed Apache-2.0. It adds 24 tokens to every session and 1,313 once invoked, about $0.0001 per session on Opus 5.5. 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 skills, from other repositories
mem0-oss-to-platform
Plan, then execute, a migration of a project from the mem0 open-source / self-hosted SDK (local Memory class) to the mem0 Platform / hosted SDK (MemoryClient). TRIGGER when: the developer wants to move, switch, or migrate mem0 off OSS/self-hosted to the hosted API, e.g. "migrate my mem0 setup to the platform", "switch…
forget
Remove a personal fact, preference, relationship detail, topic, or prior event from Muse's active memory and stop existing copies or automations from bringing it back. Use for explicit requests such as 'forget that', 'don't remember this about me', or 'remove that from your memory'. Do not use when 'forget it' merely…
Cortex
Operate Cortex, the LifeOS memory system — the typed Knowledge Archive (People, Companies, Ideas, Research with typed related: links) plus recall of prior work sessions, ISAs, and conversations. Search, add, harvest, develop, ingest, distill, graph-navigate, recall. USE WHEN cortex, knowledge, knowledge base, search…
memory
Use when the user asks to remember, recall, forget, update, search, or inspect durable OpenSquilla memory, including profile facts in USER.md and long-term notes in MEMORY.md or memory//.md.
ha-data-stores
Map of Hope Agent's local data stores and safe read-only query workflow. Use when the user asks where Hope Agent stores data, wants to inspect sessions/messages/memory/logs/background jobs/knowledge indexes/settings, asks the model to query local app data, or debugging requires checking persisted state. Trigger…
establishing-project-context
Use when the user asks to establish shared project language, or project work exposes a conflicting, renamed, or deprecated domain term that needs active semantic modeling. Routine small tasks stay on the fast path.