Stats
Actions
Tags
From engram
Manage engram plugin settings — redundancy thresholds, notification preferences, pricing, and auto-report configuration
How this skill is triggered — by the user, by Claude, or both
Slash command
/engram:configThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Manage engram plugin settings stored in `.engram/config.yaml`.
Manage engram plugin settings stored in .engram/config.yaml.
redundancy_threshold: 10000
notify_on_redundancy: true
auto_report_on_exit: false
pricing:
model: "claude-sonnet-4-20250514"
input_per_1k: 0.003
Check if .engram/config.yaml exists in the project root.
Present these options:
For each setting, show the current value and ask if they want to change it. Accept Enter to keep the current value.
redundancy_threshold (integer, default: 10000)
Number of redundant tokens detected before a notification is surfaced. Higher values reduce noise; lower values increase sensitivity.
notify_on_redundancy (true/false, default: true)
Whether to surface a notification when the redundancy threshold is crossed during a session.
auto_report_on_exit (true/false, default: false)
Whether to automatically generate a savings report when the session ends.
pricing.model (string, default: "claude-sonnet-4-20250514")
The model name used for cost estimation in reports. Must match an Anthropic model identifier.
pricing.input_per_1k (number, default: 0.003)
Cost per 1,000 input tokens in USD, used to calculate estimated savings.
After collecting all changes, display a summary of what will be written and ask for confirmation before saving.
Show the default config and ask: "Reset .engram/config.yaml to defaults? (yes/no)"
If confirmed, write the default config to .engram/config.yaml, creating .engram/ if it does not exist.
When writing the config file:
.engram/ directory exists (create it if not)..engram/config.yaml..engram/config.yaml."npx claudepluginhub pythondatascrape/engram-ccode --plugin engramCreates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.