Run a memory-consolidation pass ("dream") for a Claude Code subagent that uses the three-tier memory architecture. The agent re-reads its own .agents/ /memory/ files, decides per-entry to keep/drop/merge/promote/ enhance, applies the changes, and appends a structured stats entry to memory/DREAMS.md covering counts before/after, what was dropped and why, what was merged, what was promoted episodic→semantic, and what was rescored.
EnglishShip any HTML artifact (architecture doc, design proposal, decision card, periodic-note conversation page, code-review summary) wired to receive two complementary channels of user feedback in a single envelope: (1) explicit form elements that you embed inline — textareas, radio groups, checkboxes, action buttons — keyed by stable question IDs so you can ask the user specific things; and (2) a freeform annotation overlay — highlight-then-comment on any text, drop pinned notes on diagrams/tables/callouts, free-form sketch on top of anything, plus un-anchored "general" comments. Both channels surface live in a single right-side panel and submit together. A small stdlib Python server receives the envelope and writes a paste-ready Markdown prompt you can hand straight to the next Claude turn.
English
AGENT SKILLS ATLASIncludes Agent:dream and Annotated Feedback.
Claude Code
Codex
Cursor
GitHub Copilot
Gemini CLISupported by the Atlas install prompt. Skill-specific requirements vary.
Pack stats
What’s included
Agent:dream
Run a memory-consolidation pass ("dream") for a Claude Code subagent that uses the three-tier memory architecture. The agent re-reads its own .agents/ /memory/ files, decides per-entry to keep/drop/merge/promote/ enhance, applies the changes, and appends a structured stats entry to memory/DREAMS.md covering counts before/after, what was dropped and why, what was merged, what was promoted episodic→semantic, and what was rescored.
Annotated Feedback
Ship any HTML artifact (architecture doc, design proposal, decision card, periodic-note conversation page, code-review summary) wired to receive two complementary channels of user feedback in a single envelope: (1) explicit form elements that you embed inline — textareas, radio groups, checkboxes, action buttons — keyed by stable question IDs so you can ask the user specific things; and (2) a freeform annotation overlay — highlight-then-comment on any text, drop pinned notes on diagrams/tables/callouts, free-form sketch on top of anything, plus un-anchored "general" comments. Both channels surface live in a single right-side panel and submit together. A small stdlib Python server receives the envelope and writes a paste-ready Markdown prompt you can hand straight to the next Claude turn.
Quick start
- Choose your skills. Select individual skills above, or choose the entire pack.
- Copy the install prompt. Use Copy install prompt above the list.
- Send it to your agent. Paste into your coding agent’s chat. Your agent reviews the source and guides the setup.
Try it with your agent
After installation, start with a request like:
Use the Agent:dream skill to help with my task.
Ask me for the context you need before getting started.Requirements & license
- A coding agent that can access your project files and install skills.
- Your agent checks each skill’s instructions, dependencies, and permissions during setup.
- Reported license for the skills shown: MIT. The author’s terms apply.