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/agentsystemlabs/nebula/nebula-memorynpx skills add AgentSystemLabs/nebula --skill nebula-memorygit clone --depth 1 https://github.com/AgentSystemLabs/nebulaWrote 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/agentsystemlabs/nebula/nebula-memory)<a href="https://agentmods.dev/skills/agentsystemlabs/nebula/nebula-memory"><img src="https://agentmods.dev/badge/skills/agentsystemlabs/nebula/nebula-memory.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 | $0.00136 | $0.02311 |
| Opus 5 | $0.00068 | $0.01156 |
| Sonnet 5 | $0.00027 | $0.00462 |
| Haiku 4.5 | $0.00014 | $0.00231 |
Grade B, and why
nebula-memory scanned grade B 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
Before writing, read the index and grep the entries (`grep -ril '<symbol or TERM>' .claude/memory/entries`). How it starts
The opening of the file, as written. The whole thing — 137 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Nebula keeps a shared, committed MEMORY LOG. It has three layers, and only the first two are read by every session — the third is fetched when it matters:
| Layer | File | Who reads it | Cap |
|---|---|---|---|
| Index | .claude/MEMORY.md — one line per entry, newest first |
every session, in full | 200 lines |
| Standing gotchas | .claude/memory/gotchas.md — the durable traps, grouped by TERM |
every session, in full | 300 lines |
| Entries | .claude/memory/entries/YYYY-MM-DD-<slug>.md — the full Asked / Did / Gotchas |
the RECALL HOOK injects the matching ones per prompt; agents open the rest by index line | none |
Index lines that fall off the bottom move to .claude/memory/archive.md, which the RECALL HOOK still
searches. make memory-check (part of make ci) fails the build when a cap is exceeded or the index
and the entry files disagree, so the caps are the rule, not a request.
Your job here is the write: one entry file, one index line, and — when the task left a durable trap behind — a line in the standing gotchas.
When to write an entry
Write one when the task is done and it left something durable behind:
- code changed, or behavior changed
- a bug was diagnosed — especially if the cause was not where the symptom was
- you hit something surprising: a flaky test, a platform quirk, an agent CLI that lies about its state, a build step that has to happen in a particular order
- the user made a decision worth not relitigating ("we're not doing X because Y")
Do not write an entry for: pure questions you answered without changing anything, trivial one-line edits with no surprise in them, or anything the code and git log already say plainly. An entry that restates a diff costs every future retrieval tokens and teaches nothing.
If you found nothing durable, say so and skip the write. That is a valid outcome.
1. The entry file
Create .claude/memory/entries/YYYY-MM-DD-<slug>.md — the date from date +%F (do not guess it), the
slug the title in lower-kebab-case, at most ~60 characters. Never overwrite another entry's file.
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 · 137 lines · 136 tokens per session scan B 7d4f72f559a8
nebula-memory is a skill published in the GitHub repository AgentSystemLabs/nebula (103 stars, last pushed 3d ago), licensed MIT. It adds 136 tokens to every session and 2,311 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). 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 and then execute a migration of a project from the mem0 open-source / self-hosted SDK (the local Memory class) to the mem0 Platform / hosted / managed SDK (the MemoryClient class). Use this whenever a developer wants to move, switch, or migrate their mem0 usage off OSS/self-hosted to the hosted API — e.g.…
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…
auditing-subgroup-fairness
Audit an OpenMed NER or de-identification model for performance disparities across demographic subgroups (sex, age band, race/ethnicity when available) using openmed.eval.fairnessreport. Use when the user wants per-subgroup recall and leakage, wants to check whether de-identification under-protects a group, wants to…
agent-memory
../../../engineering/agent-memory/skills/agent-memory/SKILL.md.
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…