investigate
Find out why something failed or behaves the way it does, using read-only commands only, and report the cause back in chat.
Skill by MarvinJWendt
Chalk Labs
Need help deploying investigate?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
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/marvinjwendt-investigate
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
Use cases: documents
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 and code. Checked Nov 1, 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/investigate
What's inside
- Skill name
investigate- Description
- Find out why something failed or behaves the way it does, using read-only commands only, and report the cause back in chat.
- License
- MIT
- Instructions
- 18 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Investigate The user wants an explanation, not a fix. Find the cause and report it in chat. Editing files, applying a patch or fixing it on the way is a separate task they start themselves. Every command you run only reads: `cat`, `grep`, `git log`, `git diff`, `journalctl`, `docker logs`, `kubectl get/describe/logs`, a `--dry-run`, a `--help`. Anything that writes, installs, restarts, deletes or touches remote state waits for the user to ask for it. If the investigation dead-ends without such a command, name the command and let him decide. Dig until you have the actual cause. "The build failed" is the symptom, "the lockfile pins X 2.0 while the code calls the 1.x API" is the cause. Follow the chain backwards: the error, the code that raised it, the state that got it there, the change that created that state. A guess stays labelled as a guess. ## The report Run the `unslop` skill over it. Short and concrete, with enough detail that the user can act on it without a follow-up question. - Lead with the cause, one or two sentences. - Then the chain that proves it, carrying its evidence: file paths with line numbers, log lines, commit SHAs, command output. Quote the parts that make the point. - Name what stayed unclear and why. - Close with what would fix it, a few lines. Whether you do it is the user's call.
From MarvinJWendt/skills, MIT.
More from MarvinJWendt/skills
- 0.587score
set-up-skills
MarvinJWendt
Wire a repository's skills so `.agents/skills/` is the single source of truth and every agent reads it through a symlink.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
set-up-agents-md
MarvinJWendt
Make `AGENTS.md` the single source of truth for a repository's agent instructions and symlink every agent's own instruction file to it.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
writing-comments
MarvinJWendt
How code comments get written: what earns one, how it reads, the sediment to keep out, and the passes it goes through.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
technical-writing
MarvinJWendt
Layered technical-writing standard: text size and scope, Diátaxis structure, Google developer style sentences, STE instruction rules, Global English syntax.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
commit
MarvinJWendt
Commit the current changes with a message written in the repo's own style.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
orchestrate
MarvinJWendt
Act as team lead for a batch of parallel tasks, delegate each to a named subagent, then integrate their work and report back.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
quick-html
MarvinJWendt
Build a single self-contained HTML file with Tailwind and CDN imports, no build step.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
unslop
MarvinJWendt
Cut AI tells from prose and give it a human voice.
SkillAFreshness: A★ 1Claude Code +9 - 0.582score
set-up-agents
MarvinJWendt
Set up a repository's whole agent configuration in one go, instructions and skills.
SkillAFreshness: A★ 1Claude Code +9 - 0.582score
handoff
MarvinJWendt
Write a handoff document that summarises the current conversation, so a fresh agent can continue the work.
SkillAFreshness: A★ 1Claude Code +9 - 0.581score
pull
MarvinJWendt
Bring the current branch up to date by merging its upstream, keeping uncommitted work and both sides of every conflict.
SkillAFreshness: A★ 1Claude Code +9 - 0.579score
push
MarvinJWendt
Push the current branch to its remote, committing first when the working tree is dirty.
SkillAFreshness: A★ 1Claude Code +9
Signals
- Stars
- 1
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Oct 10, 2026 (27 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/MarvinJWendt/skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from MarvinJWendt
- 0.587score
set-up-agents-md
MarvinJWendt
Make `AGENTS.md` the single source of truth for a repository's agent instructions and symlink every agent's own instruction file to it.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
set-up-skills
MarvinJWendt
Wire a repository's skills so `.agents/skills/` is the single source of truth and every agent reads it through a symlink.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
technical-writing
MarvinJWendt
Layered technical-writing standard: text size and scope, Diátaxis structure, Google developer style sentences, STE instruction rules, Global English syntax.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
writing-comments
MarvinJWendt
How code comments get written: what earns one, how it reads, the sediment to keep out, and the passes it goes through.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
commit
MarvinJWendt
Commit the current changes with a message written in the repo's own style.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
orchestrate
MarvinJWendt
Act as team lead for a batch of parallel tasks, delegate each to a named subagent, then integrate their work and report back.
SkillAFreshness: A★ 1Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get investigate running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
investigate skill. Agent Mart. https://agentmart.dev/skills/marvinjwendt-investigate (last updated 2026-10-24)