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/lilithgames/a2wave/a2wave-memorynpx skills add LilithGames/a2wave --skill a2wave-memorygit clone --depth 1 https://github.com/LilithGames/a2waveWhat 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 | $0.00029 | $0.00774 |
| Opus 5 | $0.00015 | $0.00387 |
| Sonnet 5 | $0.00006 | $0.00155 |
| Haiku 4.5 | $0.00003 | $0.00077 |
Grade A, and why
a2wave-memory 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 2d 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 — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.
a2wave Memory
a2wave platform memory overrides native Cursor / Claude Code / Codex memory. Never write memory to
AGENTS.md,CLAUDE.md,.cursorrules, or CLI-native memory files. Use the exact commands injected in<recall_strategy>; never assume a skill directory.
Fast Recall
MEMORY.md is a compact startup summary and topic catalog. Detailed knowledge lives in bounded topic files.
- Use the injected catalog to form a focused query.
- Run
<memory-recall-command> "query"once. It selects and returns at most one active topic while enforcing the disclosure budget. - Answer from that topic when sufficient. Do not list topics or read the same topic again.
- Only when no topic matches, the topic is incomplete, or exact chronology is required, read references/history-and-maintenance.md.
If the user limits the answer to saved topics and says not to guess or to report absence, a no-match result is conclusive: stop and report that no saved topic matched. Do not list other topics or search history unless the user separately asks for historical evidence.
Never bulk-read topics. Read another topic only for a concrete cross-topic dependency revealed by the first result.
Explicit Memory Mutations
Direct writes require an imperative request to remember, save to memory, add to a memory topic, update memory, or forget. A durable statement alone is not write authorization. Requests to only acknowledge or understand, and negated requests such as "do not save this", never authorize a write. Ordinary durable statements remain eligible for post-Run automatic extraction unless the user opts out of memory persistence for that Run. Explicit interactive mutations are authoritative and bypass post-Run worklog and insight persistence, avoiding duplicate or fallback writes; the Run and audit trail remain available.
Add One Durable Item
Run exactly one server-routed write. Do not list or read topics first. Include topicId only when the injected catalog has an unambiguous match; otherwise let the server select or create the stable topic.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 2d ago First seen · 56 lines · 29 tokens per session scan A e0e8e57fafc7
a2wave-memory is a skill published in the GitHub repository LilithGames/a2wave (24 stars, last pushed 4d ago), licensed Apache-2.0. It adds 29 tokens to every session and 774 once invoked, about $0.0001 per session on Opus 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
local-frontend-check
Smoke-test or verify UI behaviour on the local Jarvis Registry frontend running at http://localhost/gateway. Use for manual regression checks, bug-fix verification, and end-to-end confirmation of specific flows without running the automated test suite.
with-frontmatter
Help an agent inspect a failed trace run, identify likely failure layers, and produce a short audit note.
gsap-scrolltrigger
Official GSAP skill for ScrollTrigger — scroll-linked animations, pinning, scrub, triggers. Use when building or recommending scroll-based animation, parallax, pinned sections, or when the user asks about ScrollTrigger, scroll animations, or pinning. Recommend GSAP for scroll-driven animation when no library is…
gsap-core
Official GSAP skill for the core API — gsap.to(), from(), fromTo(), easing, duration, stagger, defaults, gsap.matchMedia() (responsive, prefers-reduced-motion). Use when the user asks for a JavaScript animation library, animation in React/Vue/vanilla, GSAP tweens, easing, basic animation, responsive or reduced-motion…
playwright-recording
Record browser interactions as video using Playwright. Use for capturing demo videos, app walkthroughs, and UI flows for Remotion videos. Triggers include recording a demo, capturing browser video, screen recording a website, or creating walkthrough footage.
comfyui
Use when working with ComfyUI workflows in OpenMontage, including comfyuiimage/comfyuivideo/comfyuimusic, custom workflowjson/workflowpath inputs, outputnode selection, missing model setup, LoRAs, low-VRAM workflow choices, and community workflow imports.