[Feature Request] Please add Per-Agent memory panel — view/edit skill files in chat
Maintainers usually reply within 1 day
Nobody has claimed this yet.
Assessment
- Difficulty
- 5/5
- Estimated time
- Over a week
- Newbie friendliness
- 38/100
- Issue type
- Feature
- Clarity
- Mostly clear
- Activity status
- Active
- Tech stack
- typescript
- Domain
- frontend
Research direction
Start at the chat interface and trace how the selected Agent, skill or instruction file, and existing audit trail are represented. Define the panel, inline editing, validation, version history, access controls, and next-run application together; done means all seven acceptance criteria work without changing grants or enabling arbitrary code execution.
Written by the indexing model from the issue text.
Description
Problem
OpenBot has persistent coworkers, per-Agent computers, governed tool use, and auditability, but it lacks a visible, editable panel for an Agent's skill/instruction files alongside the chat interface.
For users managing multiple Agents with different roles (e.g., newsletter orchestrator, site monitor, research assistant), there is no way to quickly view or edit an Agent's skill file (e.g., SKILL.md-equivalent config) without leaving the chat or editing YAML directly. This makes it harder to refine Agent behavior iteratively or audit what instructions an Agent is operating under.
Feature request
Add Per-Agent memory panel — a viewable, editable panel next to the chat that surfaces an Agent's skill/instruction file. This should be a UI-layer feature, not a new learning infrastructure.
Proposed behaviour
- Show a collapsible panel next to the chat displaying the Agent's current skill/instruction file.
- Allow editing the skill file inline, with syntax highlighting for Markdown/YAML.
- Save edits immediately and apply them to the next run (no reload required).
- Show version history (last N edits) with timestamps and author.
- Support optional diff view to compare current vs. previous versions.
- Preserve all governance settings (grants, policies) — editing the skill file does not change grants.
Required controls
- Disabled by default; enabled per Agent.
- Panel scope: per-Agent, visible only to users with access to that Agent.
- Immutable audit log showing when a skill file was viewed, edited, or reverted.
- Review, edit, and revert controls for skill file changes.
- Clear separation between skill file content and runtime state (no arbitrary code execution from edits).
- Never automatically grant additional tools or permissions on edit — grants are unchanged.
- Configurable version history retention (e.g., keep last 50 edits).
Why this matters
This would make OpenBot more competitive for users managing multiple Agents with distinct roles while preserving its core advantages: per-Agent isolation, gateway policy enforcement, and action-level auditability. It would be particularly useful for a self-hosted team that wants to refine Agent instructions iteratively (e.g., "add this checklist to the newsletter Agent") without recreating the entire configuration.
Acceptance criteria
- A user can view an Agent's skill/instruction file in a panel next to the chat.
- A user can edit the skill file inline and save changes.
- Edits are applied to the next run without requiring a reload.
- Version history shows the last N edits with timestamps and authors.
- All skill file views/edits appear in the existing audit trail.
- Invalid skill file edits are rejected with clear error messages.
- Edited skill files respect the Agent's existing grant boundaries.
Suggested name
Agent Memory Panel or Editable Skill Files
- Dominant language
- TypeScript
- Stars
- 6.6k
- Forks
- 914
- Avg merge
- 17h 52m
- Merged PRs (30d)
- 155
Getting set up
- Ships a Dockerfile or Docker Compose file
- Has a pull request template
- No contributing guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from CopilotKit/OpenBot
-
Wrong key location at startupPossibly taken @charan-rathore claimed this today. Open
Difficulty 1/5 Under an hour Newbie friendliness 68/100
CopilotKit/OpenBot#800 ·
Maintainers usually reply within 1 day
-
setup-learning.ts fails on Windows with "spawn npx.cmd EINVAL"Possibly taken @Caldas claimed this 3 days ago. Open
Difficulty 2/5 1-3 hours Newbie friendliness 80/100
CopilotKit/OpenBot#792 ·
Maintainers usually reply within 1 day
-
Difficulty 3/5 1-2 days Newbie friendliness 68/100
CopilotKit/OpenBot#804 ·
Maintainers usually reply within 1 day
-
Five out-of-subject findings from PR #770's tier-3 review (credential-revoking tests, single-user boot gate, OAuth client rotation, shared-use atomicity, untyped server/tests)Possibly taken A pull request linked to this issue is open or already merged. Open
Difficulty 4/5 Over a week Newbie friendliness 22/100
CopilotKit/OpenBot#786 ·
Maintainers usually reply within 1 day
-
server/tests leaks database connections: 55 files open a pool and never close itPossibly taken @adamsiwiec1 claimed this 2 days ago. Open
Difficulty 3/5 1-2 days Newbie friendliness 70/100
CopilotKit/OpenBot#785 ·
Maintainers usually reply within 1 day
All issues in CopilotKit/OpenBot
Similar issues
-
Difficulty 2/5 1-3 hours Newbie friendliness 65/100
Maintainers usually reply within 1 day
-
auth: agent-bearer --manager-call is told to use the MCP instance argument on an ambiguous managerOpen
Difficulty 2/5 1-3 hours Newbie friendliness 80/100
Maintainers usually reply within 1 day
-
bug module: playback
Difficulty 2/5 1-3 hours Newbie friendliness 68/100
chthollyphile/folia-major#739 · 2 comments ·
Maintainers usually reply within 1 day
-
Difficulty 1/5 Under an hour Newbie friendliness 90/100
-
Difficulty 2/5 1-3 hours Newbie friendliness 78/100
agilepathway/label-checker#773 · 2 comments ·
Maintainers usually reply within 1 day