feat(mem0-plugin): auto coding categories, global search, OpenCode parity (#5300)
This commit is contained in:
Kartik
@@ -128,58 +128,38 @@ If `add_memory` calls fail, print:
|
||||
- Project file import failed. Check API key and MCP connection, then retry with: /mem0:onboard
|
||||
```
|
||||
|
||||
## Step 5: Set up coding categories
|
||||
## Step 5: Verify coding categories
|
||||
|
||||
Ask: "Install coding categories optimized for development workflows? [Y/n]"
|
||||
Coding categories are now configured automatically in the background when the plugin starts. This step only verifies they are set up.
|
||||
|
||||
If the user says no or skips, print `- Coding categories skipped.` and proceed to Step 6.
|
||||
Check if the categories are already configured by searching for a project_profile memory:
|
||||
|
||||
If yes, store a project profile memory that records the project and its active coding categories. Call `add_memory` once with:
|
||||
Call `search_memories` with `query="coding categories project profile"`, `filters={"AND": [{"user_id": "<active_user_id>"}, {"app_id": "<active_project_id>"}]}`, `top_k=1`.
|
||||
|
||||
- `data`: a plain-text description of the project profile and the list of active coding categories (see below)
|
||||
If a project_profile memory is found, print:
|
||||
```
|
||||
- Coding categories already configured (17 categories, auto-installed).
|
||||
```
|
||||
|
||||
If no project_profile memory is found, store one as a fallback. Call `add_memory` once with:
|
||||
|
||||
- `data`: a plain-text description of the project profile and the list of active coding categories:
|
||||
```
|
||||
Project profile for <project_id>.
|
||||
Active coding categories: architecture_decisions, api_design, data_models,
|
||||
algorithms, dependencies, environment_setup, testing_strategy, debugging_notes,
|
||||
performance, security, deployment, code_conventions, error_handling,
|
||||
refactoring_history, integrations, onboarding, project_meta.
|
||||
```
|
||||
- `user_id`: the active user id
|
||||
- `app_id`: the active project id
|
||||
- `metadata`: `{"type": "project_profile", "source": "onboard"}`
|
||||
|
||||
The standard set of 17 coding categories to include in the `data` field:
|
||||
|
||||
```
|
||||
architecture_decisions - High-level design choices and system structure
|
||||
api_design - Interface contracts, REST/GraphQL/RPC conventions
|
||||
data_models - Schemas, entity definitions, relationships
|
||||
algorithms - Non-trivial logic, performance-sensitive routines
|
||||
dependencies - Libraries, versions, upgrade notes
|
||||
environment_setup - Dev environment, tooling, build system
|
||||
testing_strategy - Test patterns, coverage targets, mocking approach
|
||||
debugging_notes - Known issues, workarounds, gotchas
|
||||
performance - Bottlenecks, profiling results, optimizations
|
||||
security - Auth patterns, secret handling, threat notes
|
||||
deployment - CI/CD pipelines, infra config, release process
|
||||
code_conventions - Naming, formatting, style rules beyond the linter
|
||||
error_handling - Error taxonomy, recovery patterns, logging approach
|
||||
refactoring_history - Past rewrites, why changes were made
|
||||
integrations - Third-party services, webhooks, external APIs
|
||||
onboarding - New-contributor notes, repo orientation
|
||||
project_meta - Goals, non-goals, stakeholder context
|
||||
```
|
||||
|
||||
Example `data` value:
|
||||
|
||||
```
|
||||
Project profile for <project_id>.
|
||||
Active coding categories: architecture_decisions, api_design, data_models,
|
||||
algorithms, dependencies, environment_setup, testing_strategy, debugging_notes,
|
||||
performance, security, deployment, code_conventions, error_handling,
|
||||
refactoring_history, integrations, onboarding, project_meta.
|
||||
```
|
||||
|
||||
After the `add_memory` call succeeds, print:
|
||||
```
|
||||
- Coding categories installed (17 categories).
|
||||
```
|
||||
|
||||
If `add_memory` fails, print the error and suggest re-running `/mem0:onboard`.
|
||||
|
||||
## Step 6: Summary
|
||||
|
||||
Print a summary:
|
||||
|
||||
@@ -1,18 +1,75 @@
|
||||
---
|
||||
name: switch-project
|
||||
description: Overrides the auto-detected project scope to read and write memories under a different project ID. Use when working across multiple projects, accessing memories from another repo, or when auto-detection resolves to the wrong project.
|
||||
description: Overrides the auto-detected project scope to read and write memories under a different project ID, or enables global search to access all memories across all users and projects. Use when working across multiple projects, accessing memories from another repo, enabling team-wide memory access, or when auto-detection resolves to the wrong project.
|
||||
---
|
||||
|
||||
# Mem0 Switch Project
|
||||
|
||||
Override the automatic project_id detection for the current directory.
|
||||
Override the automatic project_id detection for the current directory, or enable global search mode.
|
||||
|
||||
## Usage
|
||||
|
||||
The user provides a project name as an argument: `/mem0:switch-project <project-name>`
|
||||
- `/mem0:switch-project <project-name>` — switch to a specific project scope
|
||||
- `/mem0:switch-project --global` — enable global search (all memories, all users, all projects)
|
||||
- `/mem0:switch-project --no-global` — disable global search and return to per-project scoping
|
||||
|
||||
## Execution
|
||||
|
||||
### If `--global` flag is provided:
|
||||
|
||||
1. Set `global_search: true` in `~/.mem0/settings.json` using the Bash tool:
|
||||
|
||||
```bash
|
||||
python3 -c "
|
||||
import json, os
|
||||
settings_file = os.path.expanduser('~/.mem0/settings.json')
|
||||
settings = {}
|
||||
if os.path.isfile(settings_file):
|
||||
with open(settings_file) as f:
|
||||
settings = json.load(f)
|
||||
settings['global_search'] = True
|
||||
with open(settings_file, 'w') as f:
|
||||
json.dump(settings, f, indent=2)
|
||||
print('Global search enabled')
|
||||
"
|
||||
```
|
||||
|
||||
2. Print:
|
||||
```
|
||||
Global search enabled.
|
||||
Searches now return all memories across all users and projects.
|
||||
Writes still use the current user_id and app_id.
|
||||
Restart the session for the change to take effect.
|
||||
```
|
||||
|
||||
### If `--no-global` flag is provided:
|
||||
|
||||
1. Set `global_search: false` in `~/.mem0/settings.json` using the Bash tool:
|
||||
|
||||
```bash
|
||||
python3 -c "
|
||||
import json, os
|
||||
settings_file = os.path.expanduser('~/.mem0/settings.json')
|
||||
settings = {}
|
||||
if os.path.isfile(settings_file):
|
||||
with open(settings_file) as f:
|
||||
settings = json.load(f)
|
||||
settings['global_search'] = False
|
||||
with open(settings_file, 'w') as f:
|
||||
json.dump(settings, f, indent=2)
|
||||
print('Global search disabled')
|
||||
"
|
||||
```
|
||||
|
||||
2. Print:
|
||||
```
|
||||
Global search disabled.
|
||||
Searches now return only memories scoped to the current project.
|
||||
Restart the session for the change to take effect.
|
||||
```
|
||||
|
||||
### If a project name is provided (no flags):
|
||||
|
||||
1. If no project name was given, ask: "What project_id should this directory use?"
|
||||
|
||||
2. Write the mapping to `~/.mem0/project_map.json` using the Bash tool:
|
||||
|
||||
Reference in New Issue
Block a user