
Reflection
- 714 installs
- 4 repo stars
- Updated May 8, 2026
- davidkiss/smart-ai-skills
reflection is an agent skill that learns from user feedback and tool failures to update AGENT.md, CLAUDE.md, and skill definitions for developers who want their AI coding agent to improve preferences and recover from rec
About
reflection is a self-improvement skill from davidkiss/smart-ai-skills that agents must use when users give corrective feedback, request a specific working style, or when tool calls fail. It analyzes what worked, what failed, and recurring patterns, then proposes concise updates to existing skills and stores durable preferences in AGENT.md or CLAUDE.md. Developers reach for reflection when an agent keeps repeating mistakes, ignores stated conventions, or needs formalized lessons from a debugging session. The skill turns one-off chat corrections into persistent agent memory and actionable skill patches instead of losing context between sessions.
- Analyzes conversation history, tool call failures, and user corrections
- Captures explicit user preferences and project rules into AGENT.md or CLAUDE.md
- Proposes concise skill improvements with diff previews
- One change at a time with explicit user confirmation before any edit
- Hard gate: never applies changes without user approval
Reflection by the numbers
- 714 all-time installs (skills.sh)
- Ranked #1,413 of 16,546 AI & Agent Building skills by installs in the Skillselion catalog
- Security screen: MEDIUM risk (skills.sh audit)
- Data as of Jul 29, 2026 (Skillselion catalog sync)
npx skills add https://github.com/davidkiss/smart-ai-skills --skill reflectionAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 714 |
|---|---|
| repo stars | ★ 4 |
| Security audit | 3 / 3 scanners passed |
| Last updated | May 8, 2026 |
| Repository | davidkiss/smart-ai-skills ↗ |
How do coding agents learn from feedback and failures?
Systematically learn from user feedback, tool failures, and recurring patterns to continuously improve their AI coding agent.
Who is it for?
Developers operating long-running AI coding agents who need persistent preference memory and skill fixes after feedback or tool errors.
Skip if: One-shot tasks with no ongoing agent configuration, or teams that forbid modifying AGENT.md, CLAUDE.md, or local skill files.
When should I use this skill?
User provides corrective feedback, requests a specific workflow style, or a tool call fails and preferences should persist.
What you get
Updated AGENT.md or CLAUDE.md preference files and proposed improvements to existing skill definitions.
- Updated AGENT.md or CLAUDE.md preferences
- Proposed skill definition patches
Files
Reflection Skill
Overview
This skill is used to learn from interaction with the user and failures in tool calls. It analyzes what worked, what didn't (tool failures), and identifies recurring patterns or explicit user preferences that should be formalized.
Objectives
- Improve Skills: Identify gaps or inefficiencies in existing skill definitions and propose concise updates.
- Store Preferences: Capture user preferences, project-specific rules, or recurring instructions in a
AGENT.mdorCLAUDE.md(when used in Claude Code) file.
Process
1. Analyze: Review the conversation history, tool calls, and any failures or corrections from the user. 2. Identify: Determine if a specific behavior should be codified in a skill or if a user preference has emerged. 3. Propose: Formulate a single, concise change.
- If updating a skill, show a diff of the proposed change.
- If adding a preference, show the proposed addition to
CLAUDE.md.
4. Confirm: Present the proposal to the user and ask for explicit confirmation without making any changes first. 5. Apply Changes: Once user confirmed the changes, only then apply them
Guidelines
- One at a time: Only propose one change per invocation to maintain focus and allow for careful review.
- Conciseness: Keep changes as brief as possible. Often a few words are enough to clarify a requirement or fix a common mistake.
- Accuracy: Ensure the proposal directly addresses a real issue or preference observed in the session.
- Specificity: Think how you could make the learnings more generic to apply to other use cases, but don't make the changes too generic so that it would not address the original learnings
- Failure Analysis: Pay special attention to tool failures or when the user has to correct your approach. These are primary candidates for reflection.
- Conflict Resolution: If a proposed change conflicts with details of an existing skill or user preference, propose a resolution that best serves the user's current intent.
Related skills
How it compares
Use reflection for persistent agent learning; use one-off prompt edits when no AGENT.md, CLAUDE.md, or skill files should be modified.
FAQ
When must reflection be used?
reflection must be used when a user provides feedback, asks for work in a specific way, or when a tool call fails. The skill captures lessons and writes durable preferences to AGENT.md or CLAUDE.md while proposing skill improvements.
What files does reflection update?
reflection updates AGENT.md and CLAUDE.md with learned user preferences and can propose changes to existing skill definitions. This keeps agent behavior consistent across sessions instead of relying on transient chat history.
Is Reflection safe to install?
skills.sh reports 3 of 3 security scanners passed. Review the Security Audits panel on this page before installing in production.