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 Bumblebiber/hmem --skill hmem-recallgit 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-recall)<a href="https://agentmods.dev/skills/bumblebiber/hmem/hmem-recall"><img src="https://agentmods.dev/badge/skills/bumblebiber/hmem/hmem-recall/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/bumblebiber/hmem/hmem-recall"><img src="https://agentmods.dev/badge/skills/bumblebiber/hmem/hmem-recall.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00037 | $0.00432 |
| Opus 5.5 | $0.00015 | $0.00173 |
| Sonnet 5.5 | $0.00007 | $0.00086 |
| Haiku 4.5 | $0.00004 | $0.00043 |
Grade A, and why
hmem-recall 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 2mo 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.
What it actually says
hmem-recall
TRIGGER
Use when:
- You need to find past decisions, lessons, or session context from hmem
- You don't know the exact node ID
- You want to keep the search work out of the main context
STEP 1: Define the search query
Before dispatching, write down:
- QUERY: what to search for (keywords, concept, or question)
- TYPE: what kind of memory (L-Entry = lesson, O-Entry = session, P-Entry = project, any)
STEP 2: Dispatch Haiku sub-agent
Send the sub-agent exactly this prompt (fill in QUERY and TYPE):
Search hmem for: Memory type filter: <TYPE or "any">
Use these tools in order:
- search_memory(query: "") — keyword search
- find_related(id: "", query: "") — semantic search
Collect all results. Deduplicate by ID.
Return ONLY this format:
[RECALL RESULTS] | | ... [/RECALL RESULTS]
If nothing found: [RECALL RESULTS] none [/RECALL RESULTS]
Max 10 results. Most relevant first. IDs exact (e.g., L0042, O0056.3.2, P0048.6). Nothing before [RECALL RESULTS]. Nothing after [/RECALL RESULTS]. No commentary, no explanation.
STEP 3: Use results
The main agent receives the [RECALL RESULTS] block. To read a specific entry in full: call read_memory(id: "") To load a project: call load_project(id: "")
Do NOT load all results at once — pick only what the current question needs.
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.
- 2mo ago First seen · 58 lines · 37 tokens per session scan A 28c6c52aed34
hmem-recall is a skill published in the GitHub repository Bumblebiber/hmem (24 stars, last pushed 4d ago), licensed MIT. It adds 37 tokens to every session and 432 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
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.
nornweave-api
Give AI agents their own email inboxes using the NornWeave API. Use when building email agents, sending/receiving emails programmatically, managing inboxes, retrieving threads, searching messages, or integrating with email providers (Mailgun, SES, SendGrid, Resend). NornWeave provides LLM-ready markdown parsing and…
faf-expert
Expert in .faf (Foundational AI-context Format) files for persistent project context. Use when working with .faf files, project DNA, syncing project.faf into CLAUDE.md, faf-cli commands, MCP server configuration, or AI-readiness scoring (0-100%). Updated for v2.8.0 Tool Visibility System.
knowledge-base
Retrieves and updates project-specific prompt knowledge from comparison evidence and user feedback. Use only for prompt analysis or post-comparison learning within a Rashomon evaluation.
post-cortex-memory
Long-term project memory through the post-cortex MCP server (tools named session, semanticsearch, updateconversationcontext, assemblecontext, queryconversationcontext, getstructuredsummary, manageworkspace, manageentity, admin). Use it whenever those tools are available and the work could benefit from what was learned…
alice-memory
Use Alice as the user's durable local memory. Load before answering from context, and whenever you learn something worth keeping across sessions.