Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/g4rb4sh-kb
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/kb
What's inside
- Skill name
kb- Description
- Personal knowledge base of problems already solved in other projects. CHECK when starting diagnosis of a non-trivial error, an integration problem, or a technical decision that smells familiar. SAVE when the user says "save this in the kb", "this will happen to us again", or when closing a problem that was costly to solve.
- License
- MIT
- Instructions
- 30 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
Two stores, searched cheaply with grep; never open every file. | Store | What goes there | Confidentiality | | --- | --- | --- | | `docs/solutions/` in the current repo | Problems specific to this project | Stays with the client repo | | `~/.mle-setup/knowledge/` (`INDEX.md` + `cases/`) | Patterns that transfer across clients | Must be sanitized | ## Retrieve 1. `grep -ril "<2-3 symptom terms>" docs/solutions/` and `grep -i "<terms>" ~/.mle-setup/knowledge/INDEX.md`. 2. Open only the matching files. No match: continue silently. ## Record (only after the fix is verified) 1. Is it non-obvious and likely to recur? If not, skip it. 2. Draft 10-20 lines: **Symptom** (error text, how it showed up) / **Cause** / **Fix** / **Applies when** / **How to detect it quickly next time**. Tags are the words someone would grep for: error names, tools, symptoms. 3. Default: `docs/solutions/YYYY-MM-DD-<slug>.md` in the repo. If the repo's `AGENTS.md` does not mention `docs/solutions/`, propose adding one line: "Solved problems: `docs/solutions/` (grep before debugging or planning)." 4. Global case only if the lesson is useful for other clients: remove client names, data, internal URLs, architecture details, and identifiable metrics. Save `cases/YYYYMM-<slug>.md` and add `<id> | <3-5 tags> | <symptom>` to `INDEX.md` (create the folder and index if missing). 5. Show the draft to the user before saving; they confirm sanitization.
From G4rb4sh/MLE-setup, MIT.
More from G4rb4sh/MLE-setup
- 0.524score
scoping
G4rb4sh
Scoping and technical estimation of a project or feature for a client.
SkillBFreshness: BClaude Code +9 - 0.524score
handoff
G4rb4sh
Generates delivery documentation for handing a consulting project over to the client.
SkillBFreshness: BClaude Code +9 - 0.524score
llm-eval
G4rb4sh
Design or improve evaluation for an LLM-based feature (RAG, agents, classification, extraction, chat).
SkillBFreshness: BClaude Code +9 - 0.524score
pre-delivery-audit
G4rb4sh
Final audit before delivering code or a system to a client.
SkillBFreshness: BClaude Code +9 - 0.524score
repo-bootstrap
G4rb4sh
Configures a repository that does not have agent setup.
SkillBFreshness: BClaude Code +9 - 0.524score
pr-review
G4rb4sh
Review of a diff, branch, or PR before merging.
SkillBFreshness: BClaude Code +9 - 0.524score
prod-incident
G4rb4sh
Handling a production incident: something is down, degraded, or returning incorrect results in a real environment.
SkillBFreshness: BClaude Code +9 - 0.519score
ml-repro
G4rb4sh
Reproducibility checklist for ML code.
SkillBFreshness: BClaude Code +9 - 0.519score
eda
G4rb4sh
Standardized initial exploration of a new dataset.
SkillBFreshness: BClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Jul 30, 2026 (3 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/G4rb4sh/MLE-setup
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from G4rb4sh
- 0.524score
scoping
G4rb4sh
Scoping and technical estimation of a project or feature for a client.
SkillBFreshness: BClaude Code +9 - 0.524score
handoff
G4rb4sh
Generates delivery documentation for handing a consulting project over to the client.
SkillBFreshness: BClaude Code +9 - 0.524score
llm-eval
G4rb4sh
Design or improve evaluation for an LLM-based feature (RAG, agents, classification, extraction, chat).
SkillBFreshness: BClaude Code +9 - 0.524score
pre-delivery-audit
G4rb4sh
Final audit before delivering code or a system to a client.
SkillBFreshness: BClaude Code +9 - 0.524score
repo-bootstrap
G4rb4sh
Configures a repository that does not have agent setup.
SkillBFreshness: BClaude Code +9 - 0.524score
pr-review
G4rb4sh
Review of a diff, branch, or PR before merging.
SkillBFreshness: BClaude Code +9
Cite this page
kb skill. Agent Mart. https://agentmart.dev/skills/g4rb4sh-kb (last updated 2026-10-24)