Can I review the instinct before it is saved?

Absolutely. The skill generates a preview of the structured instinct—including the trigger, action, and domain—and requires user confirmation before saving.

How does Arc Recalling differ from Arc Observing?

While Arc Observing automatically detects patterns in the background, Arc Recalling is used for ad-hoc moments where a developer explicitly wants to ensure a specific insight is remembered.

Does it prevent duplicate entries in the instinct database?

Yes, the skill includes a check-duplicate function that verifies if a similar pattern has already been saved before committing a new instinct to the repository.

What fields are inferred from my natural language description?

The skill infers a kebab-case ID, a trigger condition, the specific action to take, the relevant domain, and gathers evidence from the session context.

What is the /recall command used for?

The /recall command is used to manually flag a specific insight, pattern, or technique from your current coding session so the AI can save it as a permanent 'instinct' for future use.

Arc Recalling

Name: Arc Recalling
Author: GregoryHo

byGregoryHo

0•

生产力与工作流

Captures and saves manual coding patterns and insights as reusable AI instincts during development sessions.

Arc Recalling provides a structured mechanism for developers to manually preserve valuable insights, coding patterns, and workflows directly from their active session context. By using the /recall command, users can transform fleeting observations into persistent 'instincts' that guide future AI behavior within the ArcForge ecosystem. This skill acts as a vital bridge between passive observation and deep reflection, ensuring that specific techniques—such as unique testing strategies or project-specific debugging workflows—are captured with high fidelity, verified for duplicates, and stored for long-term project memory.

主要功能

01Manual instinct capture via the /recall command

02User-confirmed preview system for high-accuracy pattern saving

03Automatic evidence extraction from current session history

04Inference of structured fields including triggers, actions, and domains

05Duplicate detection to maintain a clean instinct repository

060 GitHub stars

使用场景

01Storing project-specific architectural rules while implementing new features

02Capturing personal workflow preferences to ensure AI consistency across sessions

03Preserving a specific debugging technique discovered during a complex issue resolution

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add gregoryho/arcforge arc-recalling

For use in Claude.ai and ChatGPT

Download Skill