Config
Skill that gives AI coding agents persistent memory and quality enforcementFrom the repository description
npx -y skills add mattjaikaran/meridian --skill configAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
- runs commandsInstructs the agent to run 7 commands, including `PYTHONPATH=$MERIDIAN_HOME uv run --project $MERIDIAN_HOME -- python -c " import json from scripts.db import open_project from scripts.state import list_settings from scripts.model_profiles import get_` and 6 more.
SKILL.md
3.3 KB, 797 tokens by cl100k_base, as published. Nobody here has run it
/meridian:config — Workflow Configuration
View and modify Meridian workflow preferences. Settings are stored in the SQLite settings table and respected by plan, execute, and review commands.
Arguments
- (no args) — Show current settings.
set <key> <value>— Set a configuration value.reset— Restore all settings to defaults.profile <quality|balanced|budget|inherit>— Shortcut to set model profile.
Keywords
config, settings, preferences, profile, model, configure, setup
Available Settings
| Key | Values | Default | Description |
|---|---|---|---|
model_profile | quality, balanced, budget, inherit | balanced | Model assignment per agent type |
discuss_mode | interactive, auto, batch | interactive | Default discuss phase mode |
auto_advance | true, false | false | Auto-advance after phase completion |
interactive_execute | true, false | false | Pause after each plan for review |
tdd_required | true, false | true | Require TDD in execution |
cross_model_review | true, false | false | Use secondary AI for review |
rtk_enabled | true, false | true | Use RTK proxy for gate commands when installed |
Procedure
- Show current settings:
PYTHONPATH=$MERIDIAN_HOME uv run --project $MERIDIAN_HOME -- python -c "
import json
from scripts.db import open_project
from scripts.state import list_settings
from scripts.model_profiles import get_profile_table, format_profile_display
with open_project('.') as conn:
settings = list_settings(conn)
profile = get_profile_table(conn)
print('## Current Settings\n')
if settings:
print('| Key | Value |')
print('|-----|-------|')
for s in settings:
print(f'| {s[\"key\"]} | {s[\"value\"]} |')
else:
print('No custom settings. Using defaults.')
print()
print(format_profile_display(profile))
"
- Set a value:
PYTHONPATH=$MERIDIAN_HOME uv run --project $MERIDIAN_HOME -- python -c "
import sys
from scripts.db import open_project
from scripts.state import set_setting
key, value = sys.argv[1], sys.argv[2]
with open_project('.') as conn:
set_setting(conn, key, value)
print(f'Set {key} = {value}')
" "$KEY" "$VALUE"
- Set model profile (shortcut):
PYTHONPATH=$MERIDIAN_HOME uv run --project $MERIDIAN_HOME -- python -c "
import sys
from scripts.db import open_project
from scripts.model_profiles import set_active_profile, format_profile_display
profile = sys.argv[1]
with open_project('.') as conn:
result = set_active_profile(conn, profile)
print(format_profile_display(result))
" "$PROFILE"
- RTK token optimization:
meridian config rtk status # Show RTK install state and enabled flag
meridian config rtk enable # Enable RTK for gate commands in this project
meridian config rtk disable # Disable RTK
- Reset to defaults:
PYTHONPATH=$MERIDIAN_HOME uv run --project $MERIDIAN_HOME -- python -c "
from scripts.db import open_project
with open_project('.') as conn:
conn.execute('DELETE FROM settings WHERE project_id = ?', ('default',))
print('All settings reset to defaults.')
"
Output
Show settings as a formatted table with current values. For profile changes, display the full agent-to-model mapping.
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.