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/addxai/reflectworld/reflectworldnpx skills add addxai/ReflectWorld --skill reflectworldgit clone --depth 1 https://github.com/addxai/ReflectWorldWrote 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/addxai/reflectworld/reflectworld)<a href="https://agentmods.dev/skills/addxai/reflectworld/reflectworld"><img src="https://agentmods.dev/badge/skills/addxai/reflectworld/reflectworld.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 5.1 | $0.00000 | $0.01871 |
| Opus 5 | $0.00000 | $0.00936 |
| Sonnet 5 | $0.00000 | $0.00374 |
| Haiku 4.5 | $0.00000 | $0.00187 |
Grade A, and why
reflectworld 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 5d 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 — 171 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ReflectWorld — Visual Perception + Memory for AI Agents
You have access to ReflectWorld, a visual perception and persistent memory system. It gives you eyes and memory for cameras.
Available Tools
| Tool | Purpose |
|---|---|
reflectworld_cam |
Camera control: add sources, snapshot, watch, stop, adjust sampling, switch mode, live view, recordings |
reflectworld_perception |
VLM analysis config: set scene description, focus targets, security rules, model override |
reflectworld_memory |
Memory search & storage: episodic events, semantic patterns, procedural rules |
reflectworld_act |
Automation: notifications, identify requests, rule management |
Scene Adaptation Workflow
When a user describes what they want to monitor, follow this flow:
User describes intent
→ 1. Configure perception (scene_description + focus_targets)
→ 2. Add camera source if not already connected
→ 3. Select watch params (mode, fps, resolution)
→ 4. Store procedural rules in memory for persistent behavior
→ 5. Start watch
Step-by-step
-
Configure perception first — Call
reflectworld_perceptionwithaction=configureto setscene_descriptionandfocus_targetsso the VLM knows what matters. This takes effect immediately on the next analyzed segment. -
Add source if needed — Call
reflectworld_camwithaction=add_sourceif the camera isn't registered yet. -
Choose watch parameters based on the scene:
- Slow-changing scenes (cooking, front door):
fps=1,960x540 - Security/motion:
mode=event(motion-triggered),fps=1 - Behavior analysis (poker, games):
mode=stream,fps=0.5 - Fast action (sports, pets playing):
fps=2-5
- Slow-changing scenes (cooking, front door):
-
Store procedural rules — Call
reflectworld_memorywithaction=add_note,memory_type=proceduralfor any persistent preferences. Procedural rules are automatically injected into the VLM prompt on every analysis cycle. -
Start watch — Call
reflectworld_camwithaction=watch.
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.
- 5d ago First seen · 171 lines · 0 tokens per session scan A 3f6dda8ecd4f
reflectworld is a skill published in the GitHub repository addxai/ReflectWorld (98 stars, last pushed 1mo ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 1,871 tokens. 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
ama-memory
Use AMA memory in OpenClaw to recall prior context, capture turns, inspect stored state, end sessions, or delete a user's memory when asked.
memx
Use when an agent needs to recall, save, inspect, or audit memX memories through the bundled MCP tools.
tencentdb-agent-memory
面向 Agent 团队的长期记忆基础设施。把"对话、文档、代码"沉淀为四类可复用记忆资产： Chat Memory（L0 对话 → L1 原子 → L2 场景 → L3 人格逐层沉淀）、Skill（带版本/资源/ 触发边界的可复用技能）、LLM-Wiki（文档知识）、CodeGraph（代码图谱）。部署形态为 memory-core + memory-hub + memory-proxy 三件套 + SDK，接入 OpenClaw / Hermes / Claude Code / CodeBuddy 等。当前版本 v2.0.0。.
ama-memory
Use AMA memory in OpenClaw to recall prior context, capture turns, inspect stored state, end sessions, or delete a user's memory when asked.
mem0-dream
Consolidates stored memories by merging duplicates, resolving contradictions, and pruning stale entries. Use when memory count is high, search results feel noisy or repetitive, or periodic cleanup is needed to maintain memory quality.
mem0-vercel-ai-sdk
Mem0 provider for Vercel AI SDK (@mem0/vercel-ai-provider). TRIGGER when: user mentions "vercel ai sdk", "@mem0/vercel-ai-provider", "createMem0", "retrieveMemories", "addMemories", "getMemories", "searchMemories", "mem0 vercel", "AI SDK provider", "AI SDK memory", or is using generateText/streamText with mem0. Also…