Skip to content
Back to search

operator-memory

Operator Memory for Claude Code.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/aerovato--operator-memory/badge.svg)](/listings/aerovato--operator-memory)

HTML

<a href="/listings/aerovato--operator-memory"><img src="/api/listings/aerovato--operator-memory/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Operator is managed via the Operator Helper, which also provides setup and agent instructions. Install the Helper with npm:

npm install --global @aerovato/operator-helper

Or Bun:

bun add --global --minimum-release-age 0 @aerovato/operator-helper@latest

Then install the Operator adapter for your harness. Click each link for harness-specific information.

| Harness | Status | Install | | --- | --- | --- | | [Claude Code (CLI + Desktop)](docs/harnesses/claude-code.md) | ๐ŸŸข Fully Supported | `operator-helper install claude-code` | | [Codex (CLI + Desktop)](docs/harnesses/codex.md) | ๐ŸŸข Fully Supported | `operator-helper install codex` | | [OpenCode](docs/harnesses/opencode-v2.md) | ๐ŸŸข Fully Supported (V2) | `operator-helper install opencode` | | [OpenCode V1 (legacy)](docs/harnesses/opencode.md) | ๐ŸŸก Supported, Legacy | `operator-helper install opencode-v1` | | [Pi](docs/harnesses/pi.md) | ๐ŸŸข Fully Supported | `operator-helper install pi` | | [DeepSeek Harness](docs/harnesses/deepseek.md) | ๐ŸŸข Fully Supported | `operator-helper install deepseek` | | [Kiro](docs/harnesses/kiro.md) | ๐ŸŸก Supported | `operator-helper install kiro` |

Setup is a conversation with your agent. Run each command in a new conversation.

1. First time only: `/operator:user-init` โ€” set up your global user partition. 2. In each new project: `/operator:project-init` โ€” scaffold Operator, migrate existing documents, and index the repo. 3. Start a new conversation and do normal work.

When starting cold on an existing project, expect a sparse brain at first; ask the agent to write the first specifications for the modules you work on. Once those documents exist, later sessions maintain them as part of ordinary work.

<p align="center"><big><strong>New to Operator? Read the <a href="docs/getting-started.md">Getting Started guide</a>.</strong></big></p>

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json ยท Manifest ยท 23 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. {
  2. "name": "operator-memory",
  3. "description": "Operator Memory for Claude Code.",
  4. "owner": {
  5. "name": "Aerovato Research"
  6. },
  7. "plugins": [
  8. {
  9. "name": "operator",
  10. "source": {
  11. "source": "npm",
  12. "package": "@aerovato/operator-claude-code"
  13. },
  14. "displayName": "Operator Memory",
  15. "description": "Self-improving context engine for coding agents.",
  16. "author": {
  17. "name": "Aerovato Research"
  18. },
  19. "homepage": "https://github.com/aerovato/operator-memory/blob/main/docs/harnesses/claude-code.md",
  20. "repository": "https://github.com/aerovato/operator-memory"
  21. }
  22. ]
  23. }
README.md ยท README ยท 128 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. <p>
  2. <img src="docs/assets/banner.jpeg" alt="Operator Memory, the self-improving context engine for coding agents">
  3. </p>
  4. <p align="center">
  5. <a href="README.md">English</a> | <a href="README.zh-CN.md">็ฎ€ไฝ“ไธญๆ–‡</a>
  6. </p>
  7. ---
  8. ### Operator Memory: The self-improving context engine for coding agents.
  9. Operator Memory gives your agent a brain for documenting all their work. As the agent works, it automatically documents within this brain โ€” specs, decisions, standards, research, lessons. Every new session starts knowing everything the last one learned.
  10. #### Features
  11. - **Complete Context Engine** โ€” Documentation, memory, indexes, and skills within a single integrated system.
  12. - **Automatic Documentation** โ€” Specs, decisions, research, and more are automatically documented by the agent.
  13. - **Transparent Memory** โ€” Memory is stored as Markdown documents you can read, update, and delete.
  14. - **Sharable Knowledge** โ€” Track documents with Git and share knowledge with your team.
  15. - **Zero Infrastructure** โ€” No background agents, no embeddings, no vector database, no model configuration.
  16. ### Installation
  17. Operator is managed via the Operator Helper, which also provides setup and agent instructions. Install the Helper with npm:
  18. ```sh
  19. npm install --global @aerovato/operator-helper
  20. ```
  21. Or Bun:
  22. ```sh
  23. bun add --global --minimum-release-age 0 @aerovato/operator-helper@latest
  24. ```
  25. Then install the Operator adapter for your harness. Click each link for harness-specific information.
  26. | Harness | Status | Install |
  27. | --- | --- | --- |
  28. | [Claude Code (CLI + Desktop)](docs/harnesses/claude-code.md) | ๐ŸŸข Fully Supported | `operator-helper install claude-code` |
  29. | [Codex (CLI + Desktop)](docs/harnesses/codex.md) | ๐ŸŸข Fully Supported | `operator-helper install codex` |
  30. | [OpenCode](docs/harnesses/opencode-v2.md) | ๐ŸŸข Fully Supported (V2) | `operator-helper install opencode` |
  31. | [OpenCode V1 (legacy)](docs/harnesses/opencode.md) | ๐ŸŸก Supported, Legacy | `operator-helper install opencode-v1` |
  32. | [Pi](docs/harnesses/pi.md) | ๐ŸŸข Fully Supported | `operator-helper install pi` |
  33. | [DeepSeek Harness](docs/harnesses/deepseek.md) | ๐ŸŸข Fully Supported | `operator-helper install deepseek` |
  34. | [Kiro](docs/harnesses/kiro.md) | ๐ŸŸก Supported | `operator-helper install kiro` |
  35. Setup is a conversation with your agent. Run each command in a new conversation.
  36. 1. First time only: `/operator:user-init` โ€” set up your global user partition.
  37. 2. In each new project: `/operator:project-init` โ€” scaffold Operator, migrate existing documents, and index the repo.
  38. 3. Start a new conversation and do normal work.
  39. When starting cold on an existing project, expect a sparse brain at first; ask the agent to write the first specifications for the modules you work on. Once those documents exist, later sessions maintain them as part of ordinary work.
  40. <p align="center"><big><strong>New to Operator? Read the <a href="docs/getting-started.md">Getting Started guide</a>.</strong></big></p>
  41. ### How Operator Works
  42. Agents excel in a single session but forget everything the moment it ends. Future sessions waste tokens re-gathering an incomplete context: re-exploring the codebase, re-learning the architecture, re-teaching decisions and corrections.
  43. Operator gives the agent a durable workspace of Markdown, kept in three places:
  44. - `.operator/` โ€” private project knowledge, stays on your machine
  45. - `.operator-shared/` โ€” project knowledge published with the repository
  46. - `~/.operator/user/` โ€” your personal rules and knowledge, used across projects
  47. Every session runs the same loop:
  48. 1. **Consult** โ€” the agent starts from your Brain: instructions, codebase index, specs, guides.
  49. 2. **Build** โ€” the agent does normal development work, informed by that knowledge.
  50. 3. **Update** โ€” the agent records what changed: new specs, decisions, standards, lessons.
  51. When project truth changes, the agent updates the canonical file instead of adding a RAG database record. For more details, see [Architecture](docs/architecture.md).
  52. <p>
  53. <img src="docs/assets/change-the-loop.png" alt="The memory-aware agentic loop: consult the brain, build, update the brain">
  54. </p>
  55. ### Everyday Workflow
  56. 1. Give the agent normal development work.
  57. 2. The agent automatically consults existing knowledge: index for navigating code, specs for module contracts, guides for third-party integration details.
  58. 3. The agent automatically updates existing knowledge: When project truth changes, the relevant documents are automatically updated.
  59. 4. The next session continues from the updated knowledge.
  60. Sometimes agents hesitate to create, consolidate, or split documents. In that case, steer the agent towards making larger architectural decisions:
  61. - "Write a spec for this feature before implementing it."
  62. - "Record this research so we do not repeat the investigation."
  63. - "These two documents overlap. Consolidate them."
  64. - "This document is too large. Split it."
  65. - "Promote this spec to Shared so the team receives it."
  66. ### Roadmap
  67. **Operator Memory is under active development.** More features are on the way. Each will come with an explicit on/off switch in case you prefer the vanilla Operator experience.
  68. #### Brain Improvements
  69. - **Observation Engine** โ€” Learn durable user observations over time, kept separate from explicit User Instructions.
  70. - **Reliable Brain Updates** โ€” Keep specs and other Brain documents current during long conversations, instead of relying only on the agent to remember.
  71. #### Context Management
  72. - **Cache-Aware Preamble Rendering** โ€” Automatically refresh preamble when cache expires.
  73. - **Context Management** โ€” Custom strategies for compaction, pruning, and context management.
  74. - **Lossless Context Compression** โ€” Losslessly extend context via lossless context compression.
  75. ### VS Other Memory Plugins
  76. Snippet capture, RAG retrieval, and context compression all fail the same way.
  77. - [Read the comparison.](docs/comparison.md)
  78. - [Read the blog post.](https://liao.gg/blog/agents-dont-need-memory)
  79. - [Looking for benchmarks?](https://github.com/aerovato/operator-memory/issues/48#issuecomment-5979061413)
  80. ### Learn More
  81. - [Getting Started](docs/getting-started.md) - onboarding and usage guide: how Operator works, setup, daily use, directing the agent, and reviewing the Brain. Includes the command reference.
  82. - [Architecture](docs/architecture.md) - how partitions, catalogs, indexes, and deterministic context loading work.
  83. - [Harness docs](docs/harnesses/) - per-harness installation, verification, commands, updates, and troubleshooting.
  84. - [Troubleshooting](docs/troubleshooting.md) - validation, repair, and update recovery.
  85. ### License
  86. BSD 3-Clause. See [`LICENSE`](LICENSE).

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If itโ€™s helped, pass it on.