autospec-classify

Apply Phase 3.5 ctx and reasoning labels to GitHub auto-implement issues.

3|1|Updated Apr 29, 2026
One-click install
npx skills add https://github.com/berlinguyinca/autospec --skill autospec-classify
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: autospec-classify
Source: https://github.com/berlinguyinca/autospec/tree/main/skills/autospec-classify
Command: npx skills add https://github.com/berlinguyinca/autospec --skill autospec-classify

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires gh, jq, curl, git, and includes scripts (resource) components.

What problem does it solve?

autospec-classify removes the ambiguity in already-created auto-implement issues by applying the Phase 3.5 ctx/reasoning “model fit” rubric after the fact, so downstream autospec decisions remain explainable and consistently auditable.

Core Features & Use Cases

  • Retroactive model-fit labeling: Walks open auto-implement issues and assigns ctx:* and reasoning:* labels using the Phase 3.5 rubric.
  • Body annotation for traceability: Inserts or replaces a ## Model fit block in each issue body (delimited with markers for safe re-runs).
  • Optional project board assignment: With --apply-boards, maps labels to GitHub Projects via ~/.autospec/project-map.yml (no-op until the project-map reader lands).
  • Dry-run safety: Provides a labels/block preview without modifying issues when --dry-run is set.

Quick Start

Run /autospec-classify to classify every eligible open auto-implement issue and add a Model fit block.

Frequently Asked Questions about autospec-classify

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

FAQPage Schema
How do I retroactively label GitHub issues with AI model fit categories?

Retroactive model-fit labeling walks open auto-implement GitHub issues and assigns `ctx:*` and `reasoning:*` labels using a Phase 3.5 rubric, inserting an auditable Model fit block into each issue body for traceability.

Can I preview label changes to GitHub issues before applying them?

Yes, you can preview label and block changes without modifying issues by running the classification in dry-run mode, which outputs a safe preview of the exact `ctx:*` and `reasoning:*` labels that would be applied.

Do I need the GitHub CLI to automate issue labeling and project board assignment?

Yes, GitHub CLI (`gh`) access is required to read issues and apply labels. Optional project board assignment also requires a `~/.autospec/project-map.yml` file to map labels to GitHub Projects.

What is the best way to ensure consistent AI workflow labeling across existing implementation queues?

Applying a rubric-based ctx and reasoning model-fit classification ensures consistent labeling across existing implementation queues, keeping downstream autospec profile decisions explainable and auditable.

Is it safe to re-run automated labeling on GitHub issues that already have a Model fit block?

Yes, the labeling process uses idempotent per-issue label creation and in-place block replacement with safe delimiters, so re-running the classification safely updates existing issues without duplicating annotations.