Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsBlogPro
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges
  • Chrome Extension
  • Skill Manager

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Smoke

ASecurity

Smoke-test every repo task and CLI command; warnings count as failures.

10 stars
0 votes
0 copies
0 views
Added 10/4/2026
ai-agentsrustgobashdebugginggitdocumentation

Works with

terminalcli

Security Analysis

A100/100

Scanned 10/6/2026

$npx -y skills add fmind/dot --skill smoke --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Smoke?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Smoke
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/fmind-smoke/badge)](https://www.skillsdirectory.com/skills/fmind-smoke)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
Files
SKILL.md
---
name: smoke
description: "Smoke-test every repo task and CLI command; warnings count as failures."
license: MIT
metadata:
  kind: task
  author: Médéric HURIER (Fmind)
  source: github.com/fmind/dot/tree/main/skills/smoke
  created: "2026-10-04"
  updated: "2026-10-05"
---

# Smoke

Prove that everything a repository exposes still runs: each task, each CLI command, and each documented example, with a matrix the user can trust. Warnings count as failures. [quality-assurance](../quality-assurance/SKILL.md) owns risk-based campaigns over user journeys; [repository-review](../repository-review/SKILL.md) owns code findings. Fix defects only when the request includes fixing them.

## Workflow

1. **Baseline and isolation**: record `git status --short`. Tasks that format, lock, or generate files mutate the tree; with unrelated work present, run them in a snapshot per [git-worktree](../git-worktree/SKILL.md) and report their diff instead of leaving it behind.
1. **Discover**: list every entry point and read each command before running it; never classify by name alone.

   ```bash
   mise tasks --json --hidden | jq -r '.[] | select(.global | not) | [.name, ((.run // []) | join(" && "))] | @tsv'
   MISE_TASK_QUIET=false mise run -n <task>  # print the resolved command; quiet settings hide it
   jq -r '.scripts // {} | keys[]' package.json
   just --summary; make -pRrq : 2>/dev/null | grep -E '^[a-zA-Z0-9_.-]+:' | cut -d: -f1 | sort -u
   ```

   Add CLI entry points from `[project.scripts]`, `bin` fields, and `scripts/`, plus the commands quoted in `README.md` and CI workflows.
1. **Classify**: _safe_ (check, lint, test, local build, docs build), _mutating_ (format, fix, lock, generate: snapshot only), _service_ (serve, dev, watch: start under a timeout, probe health, stop it), and _external_ (deploy, release, publish, push, upgrade, login, prune, paid or networked writes). Run an external task only with explicit authority and its `--dry-run` when it has one; otherwise list it as skipped with the reason.
1. **Run tasks**: start with the repository's aggregate gate (`mise run all`, `check`, `test`), then each remaining task individually so failures stay attributable. Close stdin, bound time, keep logs, and force tasks whose cached outputs would skip them.

   ```bash
   mise run --force --timeout 10m <task> </dev/null >"$log" 2>&1; echo "exit=$?"
   ```

1. **Walk CLIs**: run `--help` on every command and subcommand recursively (Typer and Click list them under `Commands:`), then each read-only command (`status`, `list`, `show`, `doctor`, `validate`, any `--dry-run`) with representative arguments. Parse `--json` output with `jq`; a malformed document is a failure. Error paths count too: a bad argument must exit nonzero with a clear message, not a traceback.
1. **Treat warnings as failures**: scan each log for `warning`, `deprecat`, `WARN`, pytest's warnings summary, `Traceback`, and skipped tests; a warning passes only with a tracked, explained exception. Look at intentionally quiet tasks as well; quiet output must not hide a nonzero exit.
1. **Report the matrix**: entry point, class, exit code, duration, warnings, and note (fixed, skipped with reason, or failed with the first useful error line). Confirm that the tested snapshot equals the working tree after the last fix, and rerun only what a fix invalidated.

## Gotchas

- **Interactive prompts hang without a terminal**: closed stdin plus `--timeout` turns them into a recorded failure; prefer the tool's `--yes` only when its effect is safe.
- **Dependencies run twice**: `depends` re-runs prerequisites; reuse a passing result when inputs did not change instead of serial re-execution.
- **Run with the repository's default environment**: mise `[env]`, `.env`, and `CI=true` change behavior; run with the repository's defaults and note any variable you set.
- **Skipped is not passed**: a task skipped for a missing optional tool is "not run", not "passed".

## Documentation

- [mise run](https://mise.jdx.dev/cli/run.html) · [mise tasks](https://mise.jdx.dev/tasks/)
- Companion skills: [mise](../mise/SKILL.md) (task vocabulary), [cli-development](../cli-development/SKILL.md) (CLI contracts), [systematic-debugging](../systematic-debugging/SKILL.md) (root causes).

Attribution

fmindfmind
View sourceSee grades on GitHubMore from fmind →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Terse caveman voice: answer first, fluff gone, every technical fact kept. Use for /caveman, "caveman mode", "talk like caveman", "be brief", "less tokens". Stays on until "stop caveman" or "normal mode".

1100021 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

698621 votes

Writing Skills

Create and manage Claude Code skills in HASH repository following Anthropic best practices. Use when creating new skills, modifying skill-rules.json, understanding trigger patterns, working with hooks, debugging skill activation, or implementing progressive disclosure. Covers skill structure, YAML frontmatter, trigger types (keywords, intent patterns), UserPromptSubmit hook, and the 500-line rule. Includes validation and debugging with SKILL_DEBUG. Examples include rust-error-stack, cargo-dep...

3931 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3421 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Amp, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Grok Build, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

741 votes
View all in ai-agents →