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 U4CSolutions/MidMem --skill midmem-orchestratorgit clone --depth 1 https://github.com/U4CSolutions/MidMemWrote 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/u4csolutions/midmem/midmem-orchestrator)<a href="https://agentmods.dev/skills/u4csolutions/midmem/midmem-orchestrator"><img src="https://agentmods.dev/badge/skills/u4csolutions/midmem/midmem-orchestrator/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/u4csolutions/midmem/midmem-orchestrator"><img src="https://agentmods.dev/badge/skills/u4csolutions/midmem/midmem-orchestrator.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 | $0.00121 | $0.01013 |
| Opus 5.5 | $0.00048 | $0.00405 |
| Sonnet 5.5 | $0.00024 | $0.00203 |
| Haiku 4.5 | $0.00012 | $0.00101 |
Grade A, and why
midmem-orchestrator 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 14d 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 — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
MidMem Orchestrator — curation pipelines, frontier-planned + QA'd
This is the hermes-build-orchestrator loop (read that skill for the kanban CLI, the issue-resolution
decision tree, autonomy/reporting policy, and the DELEGATE-52 long-horizon defenses) specialized for
MidMem knowledge work. You (frontier) plan + QA; Hermes/qwen does the bulk mechanical work via
kanban; the QA gate is the midmem-ingest-review skill.
When to use
Multi-item knowledge-store work that's too much for one session: ingesting a batch of sources,
re-grounding/auditing existing entries, deduping, reviewing tier promotions, or verifying the vault
projection after a large change. Single recall/store/ingest = just use midmem-ops (OpenClaw) or the
midmem CLI directly; don't orchestrate.
The contract (what every curation card must respect)
state.dbis the source of truth; the vault is a deterministic, regenerable projection — never edit the vault; if it looks wrong,midmem projectrebuilds it. Curate the store, not the projection.- Grounding holds: ingest deterministically quarantines extracted concepts/claims not present in
the source. A card must not bypass it (no
MIDMEM_GROUNDING=0in curation). - Scope discipline: writes default to the agent's scope; publish cross-agent facts as
shared; never write another stack's private scope. - Tiers are earned, not asserted: promotion runs on usage/feedback (quantitative), never LLM
judgment. Curation should add
feedback/usage signals, not hand-promote. - Supersede on reingest: re-ingesting a changed source archives its prior entries — prefer re-ingest over manual edits.
Card patterns (decompose to these — small, verifiable)
- Ingest one source →
midmem ingest <path> --scope <s>; acceptance:success, grounding report shows acceptablesummaryScoreand no unexpected quarantine. - Re-ground / audit a slice → run
midmem-ingest-reviewover a topic; acceptance: report has no unresolved contradictions/ungrounded entries (or they're flagged for forget/supersede). - Dedup / cleanup → identify duplicates via query; supersede by re-ingest or
forget(soft); acceptance:briefcounts reconcile. - Vault verify →
midmem projectthen confirm projection matchesstate.dbcounts.
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.
- 14d ago Changed · +11 lines e73acd1d0fb6
- 1mo ago First seen · 51 lines · 121 tokens per session scan A 614497952331
midmem-orchestrator is a skill published in the GitHub repository U4CSolutions/MidMem (2 stars, last pushed 2d ago), licensed Apache-2.0. It adds 121 tokens to every session and 1,013 once invoked, about $0.0005 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 2026-08-31.
Other skills, from other repositories
ori-memory
Persistent agent memory with learning retrieval. Knowledge graph on markdown files — capture insights, decisions, research, and learnings during work, then retrieve them weeks or months later. Use when knowledge is too valuable to lose but too much to inject into every prompt.
lemmalog
Externalize working memory and logical state into the lemmalog Datalog engine (MCP). Use for ANY multi-step task where state should outlive one context window or span agents: long investigations, debugging sessions, audits, multi-agent searches, systematic explorations, planning with many interdependent constraints…
braindb
Memory recall and save. Use at every conversation start and whenever the user shares personal information, expertise, project context, preferences, or decisions worth remembering long-term.
braindb-agent
Persistent memory across sessions via the BrainDB agent. Use at conversation start and whenever you need to recall what you know about the user or save new information to long-term memory.
braindb-custom-profile
How to author a BrainDB custom profile — prompt add/replace fragments and an optional keyless ingestor — that shapes wiki naming/structure and feeds a custom ingestion source, with zero effect on defaults when inactive.
remnic-memory-workflow
Shared memory workflow for Claude Code agents connected to Remnic — recall before acting, observe during work, remember at the end. Trigger phrases include "what do you remember about", "save this for later", "any context from last time".