New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Self Improve

by ginlix-ai in ginlix-ai/LangAlpha

Report issues and propose fixes to improve your own capabilities when you encounter errors or limitations

Apache-2.0Auto-check passedDevelopment

Install Self Improve

skills CLI
$ npx skills add ginlix-ai/LangAlpha --skill self-improve -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install ginlix-ai/LangAlpha self-improve --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/ginlix-ai/LangAlpha.git skills-src && mkdir -p .claude/skills && cp -r skills-src/plugins/langalpha_service/skills/self-improve .claude/skills/self-improve && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
self-improve
GitHub stars
1.8k
Token cost
~1.7k tokens
SKILL.md length
731 words
Files
1
Skills in repo
37
Repo updated
First seen
Licence
Apache-2.0

At a glance

Report issues and propose fixes to improve your own capabilities when you encounter errors or limitations

  • Works in 6 steps: Clone or update: if… → Branch: cd .self-improve/langalpha &&… → Make the fix (keep it minimal and focused) → …
  • Development work in your project
  • SKILL.md covers When to Activate This Skill, Authentication, Workflow 1: Report a Bug… and Workflow 2: Propose a Fix…, plus 4 more sections
  • Calls gh, git and ruff; needs GITHUB_TOKEN and GITHUB_BOT_TOKEN

What it does

Self Improve is an agent skill from ginlix-ai/LangAlpha. Report issues and propose fixes to improve your own capabilities when you encounter errors or limitations

Its SKILL.md is about 1.7k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Development. It works with Git. The repository describes itself as: Claude Code for Financial Market. The licence is Apache-2.0.

When your agent uses it

  • Development work in your project

Example prompts

  • “/self-improve”

Requirements

  • A credential in GITHUB_TOKEN
  • A credential in GITHUB_BOT_TOKEN

Workflow steps

6 steps, taken from the first numbered list in SKILL.md.

  1. Clone or update: if .self-improve/langalpha exists, cd .self-improve/langalpha && git checkout main && git pull origin main to get latest…
  2. Branch: cd .self-improve/langalpha && git checkout main && git checkout -b bot/fix/
  3. Make the fix (keep it minimal and focused)
  4. Test: ruff check . && pytest (or relevant subset)
  5. Commit: conventional format — fix(scope): description
  6. PR

What it can do on your machine

Read from SKILL.md and the folder at commit e05bd91. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • gh
    • git
    • ruff
    • pytest

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use gh and git, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names these keys or tokens, usually read from environment variables:

    • GITHUB_TOKEN
    • GITHUB_BOT_TOKEN

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Self Improve loads about 1.7k tokens when it runs. Until then it costs about 30 tokens; SKILL.md has 731 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~30
When it runs · the whole SKILL.md, loaded when a task matches
~1.7k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from ginlix-ai/LangAlpha at commit e05bd91, republished under its Apache-2.0 licence (© ginlix-ai). 731 words, ~1,679 tokens.

Download SKILL.mdSave it as .claude/skills/self-improve/SKILL.md (or your agent's skills folder).
name
self-improve
description
Report issues and propose fixes to improve your own capabilities when you encounter errors or limitations

Self-Improvement

You have access to gh CLI and git in the sandbox. Use this skill when you encounter bugs, errors, or limitations in your own capabilities during your work.

When to Activate This Skill

Recognize these situations during your normal work:

  • A tool call fails with an unexpected error
  • A skill's instructions (SKILL.md) have wrong examples or outdated API usage
  • An MCP tool returns malformed or unexpected data
  • A prompt instruction is unclear, contradictory, or missing
  • A sandbox script or package is broken or misconfigured
  • You notice a code pattern that would make your work more effective if fixed

Always ask the user for consent before filing an issue or creating a PR. Briefly explain what you found and what you want to report, then ask for approval. Do NOT file anything without explicit user confirmation.

Do NOT derail the user's current task. Keep the consent request brief, file the issue quickly after approval, then continue with the user's request.

Authentication

  • GITHUB_TOKEN env var is pre-configured (injected from GITHUB_BOT_TOKEN)
  • Git identity is set via env vars — no git config needed

Workflow 1: Report a Bug (default — quick, no cloning)

Use when: you encounter any issue with your own capabilities. Time: <30 seconds. Don't interrupt the user's flow.

Template:

bash
gh issue create --repo "ginlix-ai/LangAlpha" \
  --title "bug(agent): <what broke>" \
  --label "agent-reported" \
  --body "<structured body>"

Issue body structure:

## What I was doing
<user's task context — what were you trying to accomplish>

## What went wrong
<exact error message or unexpected behavior>

## Where the issue likely is
<file paths, function names, skill names — be specific>

## Suggested fix
<if obvious, describe; otherwise "Needs investigation">

## Environment
- Thread: <thread_id if available>
- Tool/Skill: <which tool or skill was involved>
- Error type: <tool_error | skill_instruction | mcp_data | prompt | sandbox>

@claude Please triage this issue — verify the root cause, assess severity, and suggest a fix if straightforward.

Workflow 2: Propose a Fix (rare — only when user explicitly asks)

Default to Workflow 1 (filing an issue). Only create a PR when the user explicitly asks you to fix it yourself. Do NOT propose PRs on your own initiative.

Steps:

  1. Clone or update: if .self-improve/langalpha exists, cd .self-improve/langalpha && git checkout main && git pull origin main to get latest. Otherwise gh repo clone "ginlix-ai/LangAlpha" .self-improve/langalpha -- --depth 1
  2. Branch: cd .self-improve/langalpha && git checkout main && git checkout -b bot/fix/<short-desc>
  3. Make the fix (keep it minimal and focused)
  4. Test: ruff check . && pytest (or relevant subset)
  5. Commit: conventional format — fix(scope): description
  6. PR:
bash
gh pr create --repo "ginlix-ai/LangAlpha" \
  --base main \
  --title "fix(agent): <what's fixed>" \
  --label "agent-reported" \
  --body "<structured body>"

PR body structure:

## Problem
<link to issue if filed, or describe the bug>

## Root Cause
<what was wrong and why>

## Fix
<what was changed and why this approach>

## Testing
<what tests were run, what was verified>

## Context
- Discovered during: <brief user task description>
- Thread: <thread_id>

Codebase Guide — Where to Look

Use this to identify the right module when filing issues or proposing fixes.

DirectoryWhat lives hereExample issues
skills/Skill SKILL.md instructions and assetsWrong examples in skills/dcf-model/SKILL.md, bug in a provided script snippet, outdated API usage, missing steps in a workflow, new best practice to add
mcp_servers/MCP server implementations (yfinance, fundamentals, macro, price_data)yfinance_mcp_server.py returns malformed data, a fundamentals endpoint is missing a field, macro data has wrong units
src/tools/External tool implementations (web fetch, crawl, search, SEC, market data)fetch.py times out on certain URLs, SEC filing parser fails on 10-K amendments, search returns stale results
src/ptc_agent/agent/tools/Core sandbox tools (ExecuteCode, Bash, file ops, grep, glob, think, todo)code_execution.py mishandles large stdout, bash.py doesn't escape special chars, file_ops.py fails on binary files
src/ptc_agent/agent/middleware/Middleware stack (skills, subagents, compaction, memory, caching)Skill loading fails silently, subagent doesn't inherit context, compaction truncates important content
src/ptc_agent/agent/prompts/System prompt templates (Jinja2) and configRedundant or wrongful instructions in system.md.j2, useful tips and experience worth persisting into prompts
Show full SKILL.md (241 more words)Show less

Label Convention

  • Always use agent-reported label
  • Add bug for broken behavior, enhancement for capability gaps
  • Add scope labels: skills, tools, mcp, prompt, sandbox

Safety Rules

  • NEVER push directly to main — always bot/fix/ or bot/feat/ branches
  • main branch contains the latest code. Always branch from main, target PRs to main
  • ALWAYS run linting and tests before creating a PR
  • Keep PRs small — one fix per PR, max 1-3 files
  • Clone to .self-improve/langalpha (inside workspace, persists across restarts)
  • NEVER commit tokens, secrets, API keys, or user data
  • NEVER include confidential or private information in issues or PRs — no user data, no internal business context, no API responses containing private data, no conversation content. Describe the technical problem only.
  • After filing/PR, immediately return to the user's original task

Pre-Submit Checklist

Go through EVERY item before running gh issue create or gh pr create:

  • User consent obtained — user explicitly approved filing this issue/PR
  • No secrets or tokens — title, body, and diff contain zero credentials, API keys, or env values
  • No private data — no user names, portfolio holdings, conversation content, or internal business context
  • No raw API responses — sanitize or omit any data returned from MCP tools or external APIs
  • Technical description only — the issue/PR describes the bug or fix, not what the user was working on
  • Correct repo — targeting ginlix-ai/LangAlpha
  • Correct branch (PRs only) — branched from main, PR base is main
  • Minimal diff (PRs only) — only the files needed for the fix, no unrelated changes

© ginlix-ai, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in plugins/langalpha_service/skills/self-improve of ginlix-ai/LangAlpha.

Open the folder on GitHubat commit e05bd91

Compare with similar skills

Self Improve next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Self Improve compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Self Improve this skillginlix-ai/LangAlpha1.8k—~1.7kAutomated safety check: PassApache-2.0
Finishing a Development Branchobra/superpowers297k5 repos~1.9kAutomated safety check: PassMIT
Code Review ChecklistshareAI-lab/learn-claude-code78k5 repos~1.1kAutomated safety check: PassMIT
Code Design Rationale Investigatorcursor/plugins10k9 repos~2.6kAutomated safety check: PassNone
Contributor-First PR MergeHKUDS/OpenHarness16k1 repos~847Automated safety check: PassMIT
Finishing A Development Branchfarm-fe/farm5.6k34 repos~1.8kAutomated safety check: PassMIT

Similar skills

  • Finishing a Development Branch

    obra/superpowers

    Walks the last step of a branch: confirm tests pass, detect the git environment, ask how to integrate, carry out your choice and clean up the worktree.

    297k GitHub starsUsed in 5 repos~1.9k tokens
    DevelopmentAuto-check passed

  • Code Review Checklist

    shareAI-lab/learn-claude-code

    Reviews code against a five-part checklist covering security, correctness, performance, maintainability and testing, and reports findings in a fixed format.

    78k GitHub starsUsed in 5 repos~1.1k tokens
    DevelopmentAuto-check passed

  • Code Design Rationale Investigator

    cursor/plugins

    Official

    Digs into why code is shaped the way it is by checking git history, pull requests and connected tools in parallel, then reporting a cited read on the tradeoffs.

    10k GitHub starsUsed in 9 repos~2.6k tokens
    DevelopmentAuto-check passed

  • Contributor-First PR Merge

    HKUDS/OpenHarness

    Merges external GitHub pull requests while keeping the original author credited, and fixes conflicts after the merge instead of rewriting the contribution.

    16k GitHub starsUsed in 1 repo~847 tokens
    DevelopmentAuto-check passed

  • Finishing A Development Branch

    farm-fe/farm

    A skill your agent uses when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for…

    5.6k GitHub starsUsed in 34 repos~1.8k tokens
    DevelopmentAuto-check passed

  • Migrate Internal Package into Ghost

    TryGhost/Ghost

    Moves a package from another TryGhost repository into Ghost as an internal workspace package while keeping its Git history, with checkpoints for the steps that need an administrator.

    56k GitHub stars~3.8k tokensUpdated today
    DevelopmentAuto-check passed

More from ginlix-ai/LangAlpha

All 37 skills in this repo

  • Investment Deck Check

    ginlix-ai/LangAlpha

    Quality-checks an investment deck in .pptx form before it goes out: number consistency, chart and narrative alignment, source coverage, language and a circulation verdict.

    1.8k GitHub stars~3.7k tokensUpdated yesterday
    Auto-check passed

  • Equity Initiation Report

    ginlix-ai/LangAlpha

    Produces a first-time equity research initiation report in five tasks: company research, financial model, valuation, charts and a DOCX report.

    1.8k GitHub stars~3.8k tokensUpdated yesterday
    Auto-check passed

  • Three-Statement Model Builder

    ginlix-ai/LangAlpha

    Builds or repairs an integrated income statement, balance sheet and cash flow model in Excel with live formulas, supporting schedules, scenarios and a Checks sheet.

    1.8k GitHub stars~5.4k tokensUpdated yesterday
    Auto-check passed

  • Financial Model Checker

    ginlix-ai/LangAlpha

    Audits an existing Excel financial model without editing it, checking structure, formulas, integrity identities and source tie-out, and ends in a prioritized issue log.

    1.8k GitHub stars~4.2k tokensUpdated yesterday
    Auto-check passed

  • DCF Model Builder

    ginlix-ai/LangAlpha

    Builds a live Excel DCF valuation workbook with free cash flow projections, WACC, terminal value, three scenarios, sensitivity grids and a reverse DCF.

    1.8k GitHub stars~7.7k tokensUpdated yesterday
    Auto-check passed

  • Word Documents for Human Review

    ginlix-ai/LangAlpha

    Builds Word files with python-docx, edits existing ones in place with tracked changes and comments, then renders and validates the result.

    1.8k GitHub stars~4.8k tokensUpdated yesterday
    Auto-check passed

Works with

Categories

Questions about Self Improve

What does Self Improve do?

Report issues and propose fixes to improve your own capabilities when you encounter errors or limitations. Self Improve is an agent skill from ginlix-ai/LangAlpha.

When should I use Self Improve?

Self Improve fits situations like: development work in your project.

How do I install Self Improve in Claude Code?

Run `npx skills add ginlix-ai/LangAlpha --skill self-improve -a claude-code`. Or copy the skill folder (plugins/langalpha_service/skills/self-improve in ginlix-ai/LangAlpha) into .claude/skills/self-improve in your project. Claude Code loads it when a task matches its description.

How do I install Self Improve in Codex?

Run `npx skills add ginlix-ai/LangAlpha --skill self-improve -a codex`. Or copy the skill folder (plugins/langalpha_service/skills/self-improve in ginlix-ai/LangAlpha) into .agents/skills/self-improve in your project. Codex loads it when a task matches its description.

Can I use Self Improve in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add ginlix-ai/LangAlpha --skill self-improve -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/self-improve, .gemini/skills/self-improve, .github/skills/self-improve and .opencode/skills/self-improve in your project.

What does Self Improve need to run?

Going by SKILL.md and its folder, Self Improve needs the command-line tools its instructions call (gh, git, ruff and pytest) and credentials named GITHUB_TOKEN and GITHUB_BOT_TOKEN. Our summary lists: A credential in GITHUB_TOKEN; A credential in GITHUB_BOT_TOKEN.

Does Self Improve access the network?

SKILL.md contains no URLs. Its commands use gh and git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Self Improve safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Self Improve use?

Self Improve is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Self Improve use?

About 1.7k tokens (SKILL.md is roughly 6.7k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Self Improve?

Skills that share tags, products or a category with Self Improve: Finishing a Development Branch (obra/superpowers, 297k stars), Code Review Checklist (shareAI-lab/learn-claude-code, 78k stars), Code Design Rationale Investigator (cursor/plugins, 10k stars) and Contributor-First PR Merge (HKUDS/OpenHarness, 16k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Self Improve?

ginlix-ai (a GitHub organization) maintains it in ginlix-ai/LangAlpha, which has 1,811 GitHub stars. The repository holds 37 skills in this directory. The repository was last updated on October 9, 2026.

Source: ginlix-ai/LangAlpha on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.