backprop

Propagate session learnings to update existing skills via a subagent.

1|Updated Mar 15, 2026
One-click install
npx skills add https://github.com/hafley66/claude-research --skill backprop
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: backprop
Source: https://github.com/hafley66/claude-research/tree/main/skills/backprop
Command: npx skills add https://github.com/hafley66/claude-research --skill backprop

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Propagate learnings to improve existing skills by feeding observations, corrections, gotchas, or new patterns to a backprop subagent that decides how to update skills.

Core Features & Use Cases

  • Take a learning, correction, or observation from the current session
  • Spawn a backprop subagent that searches existing skills for relevance
  • The subagent edits the affected skills to incorporate the new knowledge
  • Reports what changed and why

When to use me

Trigger on:

  • "backprop this" / "propagate this back"
  • "this skill is wrong about X"
  • "remember that X actually works like Y"
  • "add this gotcha to the relevant skill"
  • Any request to update skills based on something discovered during work

How to invoke

The user provides a learning. It can be:

  1. A correction: "ast-grep patterns don't work with JSX fragments unless you set language to tsx"
  2. A new pattern: "found that combining has + inside with stopBy: end is the way to match deeply nested X inside Y"
  3. A gotcha/anti-pattern: "never use regex on multiline nodes, it only sees the first line"
  4. A missing concept: "the tree-sitter-queries skill doesn't cover the set! directive for injection.combined"

Workflow

When this skill loads, the main agent MUST:

  1. Collect the learning from the user's message (and any relevant context from the current conversation)
  2. Spawn the backprop subagent with a prompt containing:
    • The exact learning/correction/observation
    • Which topic area it relates to (if obvious)
    • Any code examples or evidence from the session
  3. The subagent handles everything else: searching skills, deciding edits, applying them
  4. Report the subagent's summary back to the user

Prompt template for the subagent

LEARNING:
{the observation, correction, or new pattern}

TOPIC HINT:
{topic area if known, or "unknown -- search broadly"}

EVIDENCE:
{any code snippets, error messages, or examples that support the learning}

The subagent will search ~/.agents/skills/, read relevant skills, and edit them.

What the subagent decides

For each relevant skill, the subagent picks one of:

  • append: Add to an existing section (new example, new bullet point, new gotcha)
  • correct: Fix incorrect information
  • restructure: Move content between sections or add a new section
  • skip: Skill is tangentially related but the learning doesn't belong there

The subagent does NOT:

  • Create new skills (that's what docs-to-skills is for)
  • Delete large sections
  • Change skill frontmatter (name, description) unless the description is now inaccurate

Frequently Asked Questions about backprop

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

FAQPage Schema
How do I propagate learnings to update and improve existing AI skills?

You propagate learnings by feeding a correction, gotcha, or new pattern to a backprop subagent that searches relevant skills, applies edits, and reports what changed and why.

What is skill knowledge propagation and when should I use it?

Skill knowledge propagation is the process of updating existing skills with new session discoveries. Use it when you find corrections, missing concepts, or anti-patterns that should influence related skills across topics.

How do I add a new gotcha or correction to an existing skill automatically?

Provide the gotcha or correction with supporting code evidence, and the backprop subagent will decide whether to append, correct, restructure, or skip it for relevant skills.

Can the skill propagation subagent create new skills from session observations?

No, the skill propagation subagent cannot create new skills. It only edits existing skills by appending, correcting, or restructuring content; creating new skills requires a separate docs-to-skills process.

How do I fix incorrect information across multiple AI skills at once?

To fix incorrect information across skills, submit the correction to the backprop subagent, which searches broadly for relevance and applies the necessary corrections to all affected skills automatically.

Does the skill update process modify skill frontmatter like name and description?

No, the skill update process does not modify skill frontmatter such as name or description unless the description is now inaccurate. It also does not delete large sections of existing skills.