skills.rest
brewpirate
brewpirateCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

scan

Generate model-scoped trigger phrases for files and directories.

Updated Jun 10, 2026
Search Tags:#repository#scan#confidence-scoring#semantic-analysis#trigger-phrases#researcher-agent#model-scoping
One-click install
npx skills add https://github.com/brewpirate/acme-frontier-ai --skill scan-brewpirate
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: scan
Source: https://github.com/brewpirate/acme-frontier-ai/tree/main/catalog/projects/total-recall/skills/scan
Command: npx skills add https://github.com/brewpirate/acme-frontier-ai --skill scan-brewpirate

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill turns files and directories into reliable semantic trigger phrases so agents can identify the right internal lookup handles for different models.

Core Features & Use Cases

  • Model-scoped scanning: Generates triggers tailored to the target models you specify instead of using one generic phrase.
  • Repository and file analysis: Handles either a single file or a directory of source, documentation, and config files.
  • Convergence-based results: Uses repeated agent sampling and pattern convergence to produce higher-confidence phrases.
  • Operational output: Stores model-specific trigger metadata, preserves existing results for other models, and rebuilds the master index after scanning.

Quick Start

Use the scan skill to analyze the provided file or directory for the target models you want and generate trigger phrases with confidence results.

Frequently Asked Questions about scan

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate model-specific trigger phrases for repository files?▼

This skill performs model-scoped repository scanning by sampling source, configuration, and documentation files to produce convergent semantic trigger phrases. It preserves existing results for other models and rebuilds the master index after scanning.

How does confidence scoring work for semantic trigger phrase analysis?▼

Confidence scoring for semantic analysis relies on pattern convergence from repeated agent sampling across repository files. Higher convergence during scanning produces higher confidence trigger phrases for your specified target models.

Can I scan a single configuration file instead of a full repository directory?▼

Yes, you can scan a single configuration file instead of a full repository directory. The skill handles both individual files and directories of source, documentation, and config files to generate model-scoped trigger phrases.

Does scanning new target models overwrite trigger phrases generated for previous models?▼

Scanning new target models does not overwrite previous trigger phrases. The skill preserves existing results for other models and only updates the master index by appending the new model-specific metadata using relative paths.

What is the best way to rebuild a trigger index after adding new source files?▼

The best way to rebuild a trigger index after adding new source files is to run the repository scanning task. It samples the new artifacts, generates model-specific triggers, and automatically rebuilds the master index with updated confidence reporting.

Related Skills

Explore More Education & Research Skills→
RlierRlier
1

oracle

Bundle prompts with files and directories into single AI model requests.

Community
Advanced
brewpiratebrewpirate

index

Parse .claude/triggers.json and write .claude/recall-index.json with phrase and weight fields.

Community
Intermediate
bennybennisonbennybennison

continuous-learning

Extract reusable patterns from agent sessions and codify them as framework additions.

Community
Intermediate
brewpiratebrewpirate

seed

Scan repository documentation and prompt files to generate model-scoped trigger phrases.

Community
Advanced