skills.rest
chrono-meta
chrono-metaCOMMUNITY
Explore all skills by this author →
GitHub

edit-manifest

Verify harness asset edits against declared predictions in edit_manifest.yaml.

7|Updated May 26, 2026
Search Tags:#claude code#change verification#manifest tracking#edit validation#impact tracking#predictive verification#harness management
One-click install
npx skills add https://github.com/chrono-meta/forge-harness --skill edit-manifest
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: edit-manifest
Source: https://github.com/chrono-meta/forge-harness/tree/main/plugins/fh-meta/skills/edit-manifest
Command: npx skills add https://github.com/chrono-meta/forge-harness --skill edit-manifest

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

When you edit harness assets like SKILL.md files, rules, or CLAUDE.md, it is impossible to know if the change actually delivered the expected improvement or introduced uncaught regressions. This Skill closes the forward verification loop after regression guards catch backward issues, eliminating guesswork about edit effectiveness.

Core Features & Use Cases

  • Predict-Verify Edit Loop: Records a falsifiable predicted impact for every harness edit, then verifies predictions against actual session outcomes in the next cycle.
  • Validation Gate: Only accepts edits that show measurable positive improvement, while retaining failed edits in a negative feedback buffer to prevent repeating mistakes.
  • Use Case: For teams maintaining Claude Code meta-harnesses, this ensures every change to skills, rules, or agent configurations is proven to deliver value, not just assumed to help.

Quick Start

Use the edit-manifest skill to record a predicted impact for your latest harness asset edit and schedule verification for the next session.

Frequently Asked Questions about edit-manifest

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

FAQPage Schema
How do I verify if edits to Claude Code rules and SKILL.md files actually improve the workflow?▼

Predictive verification tracks edit effectiveness by recording a falsifiable predicted impact for every harness asset edit, then verifies predictions against actual session outcomes in the next cycle. This ensures changes deliver measurable value.

What is predictive impact tracking for harness asset edits?▼

Predictive impact tracking is a verification loop that measures actual impact of harness edits against declared predictions. It appends manifest entries to edit_manifest.yaml and enforces a validation gate accepting only edits with measurable positive impact.

How do I validate changes to CLAUDE.md and prevent repeating failed edits?▼

Validate changes by scheduling verification against primary session signals in the next cycle. Failed edits are retained in a negative feedback buffer, preventing you from repeating mistakes while accepting only proven improvements.

Can I use edit-manifest to track regressions in Claude Code meta-harness configurations?▼

Yes, edit-manifest applies to all edits to SKILL.md files, rules/*.md, and CLAUDE.md within the forge-harness meta-harness. It closes the forward verification loop after regression guards catch backward issues.

When do I need manifest tracking for Claude Code workflow edits?▼

You need manifest tracking when maintaining Claude Code meta-harnesses and editing skills, rules, or agent configurations. It eliminates guesswork by proving every change delivers value rather than assuming it helps.

What files does edit-manifest track for change verification in Claude Code?▼

Edit-manifest tracks all edits to SKILL.md files, rules/*.md, and CLAUDE.md within the forge-harness meta-harness. It requires appending manifest entries to tracks/_meta/edit_manifest.yaml for verification.

Related Skills

Explore More Software Engineering Skills→
chrono-metachrono-meta
7

field-harvest

Scan git commit history to detect reusable patterns and log session work.

Community
Advanced
pasie15pasie15

scenario-analyzer

Generate an 18-month investment scenario report in Japanese from a news headline.

Community
Advanced
zizzfizzixzizzfizzix

pre-implement-spec

Analyze specifications for backward compatibility issues, risks, and gaps before coding.

Community
Advanced
china-qijizhifengchina-qijizhifeng
812

agentic-harness-engineering

Audit and validate coding agent harness structures against HARNESS.md v1.0.

Official
Advanced