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 skills/bumblebiber/hmem/hmem-readnpx skills add Bumblebiber/hmem --skill hmem-readgit clone --depth 1 https://github.com/Bumblebiber/hmemWrote 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/bumblebiber/hmem/hmem-read)<a href="https://agentmods.dev/skills/bumblebiber/hmem/hmem-read"><img src="https://agentmods.dev/badge/skills/bumblebiber/hmem/hmem-read.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 | $0.00085 | $0.03947 |
| Opus 5 | $0.00043 | $0.01973 |
| Sonnet 5 | $0.00017 | $0.00789 |
| Haiku 4.5 | $0.00009 | $0.00395 |
Grade A, and why
hmem-read scanned grade A with 1 finding 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 4d 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.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
| Old protocol entries | `.7.32 - windowsHide: true on all child_process spawns` (fragment, not a session) | `update_memory(id="P0048.7.32", content="...", irrelevant=true)` | How it starts
The opening of the file, as written. The whole thing — 413 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Load Memory — Choose Your Path
When this skill is invoked, pick the right path and execute a tool call immediately. Do NOT just read this document — act.
If read_memory is not available, tell the user:
"read_memory tool not found. Run hmem init to configure the MCP server."
Announcements: If read_memory or hmem-sync pull shows urgent announcements
(yellow warnings at the top), act on them immediately before doing anything else.
These are broadcast messages from the user or another device — typically config changes,
server migrations, or breaking updates that must be handled first.
Path A: Start Working on a Project (primary workflow)
Use load_project as the single entry point. It activates the project, returns the full
briefing (L2 content + L3 titles), and shows recent O-entries with full L4/L5 user/agent
exchanges — so you see exactly what happened in previous sessions, not just titles.
-
User mentioned a project → load it directly:
load_project(id="P0037")This replaces the old pattern of
update_memory(active=true)+read_memory(). One call gives you everything: project structure, open tasks, and recent session history. -
Unclear which project → list projects first, then load:
read_memory(titles_only=true, prefix="P") # ~200 tokens overviewThen ask the user or pick the most relevant one, and call
load_project(id="P00XX"). -
New project → create it first, then load:
write_memory(prefix="P", content="New project: ...") # → P00XX load_project(id="P00XX") -
Cross-project reference — need context from another project without switching active project:
read_project(id="P0048") # focused briefing, does NOT activateReturns Overview, Codebase titles, Usage, Context, Requirements, Roadmap titles. Omits: session history (O-entries), Bugs, History, Ideas, Next Steps, Rules, R-injection. Use this when working on Project A but needing the API/architecture of Project B.
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.
- 4d ago First seen · 413 lines · 85 tokens per session scan A 613be9580806
hmem-read is a skill published in the GitHub repository Bumblebiber/hmem (23 stars, last pushed 1mo ago), licensed MIT. It adds 85 tokens to every session and 3,947 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
kayba-ace
This skill ships learnfromtraces.py, a script that reads OpenClaw session transcripts, feeds them through the ACE learning pipeline, and writes an updated skillbook to disk.
rekal-init
Bootstrap rekal memory for a project. Scans the codebase for architecture, conventions, dependencies, workflows, and config, then stores durable knowledge as properly typed, tagged, deduplicated memories. Use when starting rekal on a new project, or when user says "init rekal", "bootstrap memory", "populate rekal"…
rekal-save
End-of-session memory capture with deduplication. Extracts durable knowledge, checks for duplicates, stores or replaces as appropriate. Use whenever a session wraps up, a task finishes, or the user says goodbye/thanks/done. Also use when significant preferences, decisions, or discoveries emerge mid-session. Make sure…
rekal-usage
Operational guide for rekal memory tools. Precise rules for when/how to call each tool, with exact parameters and decision trees. Use at session start, when onboarding to a rekal workspace, or when user asks "how do I use rekal", "what rekal tools", "help with memory". Trigger: /rekal-usage.
rekal-hygiene
Periodic memory maintenance and cleanup. Finds duplicates, contradictions, and quality issues in the memory database. Proposes fixes for user approval. Never auto-deletes or auto-modifies. Use when user says "clean up memories", "memory maintenance", "check memory health", or invokes /rekal-hygiene. Run monthly or…
vault-for-llm
Connect OpenClaw to Vault Agent Memory as a local-first governed project memory layer. Search first, then bounded-read cited source ranges; propose new memories as candidates instead of writing directly into active memory.