Run any Skill in Manus with one click

bash-agent

Stars2

Forks0

UpdatedMarch 8, 2026 at 17:18

Run bash-agent.sh — a lean shell script validation runner that produces agent-friendly output. Use when: running shell/bash script checks (syntax, lint), verifying shell scripts before committing, or when the user asks to run shellcheck, bash checks, or validate shell scripts. Triggers on: bash agent, shell agent, shellcheck, bash lint, shell checks, verify shell scripts.

Installation

Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.

Run Skill in Manus

Source

codesoda

codesoda/x-agent

View GitHub Repository View Creator Repositories

Download

Run Skill in Manus

Related occupationsSOC

Based on SOC occupation classification

Software DevelopersComputer and Mathematical Occupations·SOC 15-1252

File Explorer

2 files

SKILL.md

readonly

name	bash-agent
description	Run bash-agent.sh — a lean shell script validation runner that produces agent-friendly output. Use when: running shell/bash script checks (syntax, lint), verifying shell scripts before committing, or when the user asks to run shellcheck, bash checks, or validate shell scripts. Triggers on: bash agent, shell agent, shellcheck, bash lint, shell checks, verify shell scripts.
context	fork
allowed-tools	["Bash(scripts/bash-agent.sh)","Bash(RUN_=* scripts/bash-agent.sh)","Bash(SHELLCHECK_SEVERITY= scripts/bash-agent.sh)","Bash(MAX_LINES= scripts/bash-agent.sh)","Bash(KEEP_DIR= scripts/bash-agent.sh)","Bash(FAIL_FAST= scripts/bash-agent.sh)","Bash(CHANGED_FILES= scripts/bash-agent.sh*)"]

Bash Agent

Run the bash-agent.sh script for lean, structured shell script validation output designed for coding agents.

Script Location

scripts/bash-agent.sh

Usage

Run Full Suite (syntax + lint)

scripts/bash-agent.sh

Run Individual Steps

scripts/bash-agent.sh syntax    # bash -n syntax check only
scripts/bash-agent.sh lint      # shellcheck lint only
scripts/bash-agent.sh all       # full suite (default)

Environment Knobs

Variable	Default	Description
`RUN_SYNTAX`	`1`	Set to `0` to skip syntax check
`RUN_LINT`	`1`	Set to `0` to skip shellcheck lint
`SHELLCHECK_SEVERITY`	`warning`	shellcheck `--severity` level (`error`, `warning`, `info`, `style`)
`FAIL_FAST`	`0`	Set to `1` to stop after first failure (or use `--fail-fast`)
`CHANGED_FILES`	(empty)	Space-separated changed file paths; scopes checks to those `.sh` files
`MAX_LINES`	`40`	Max output lines printed per step (unlimited in CI)
`KEEP_DIR`	`0`	Set to `1` to keep temp log dir on success

Output Format

Each step prints a header (Step: syntax, Step: lint)
Results are PASS, FAIL, or SKIP
On failure, output is truncated to MAX_LINES
Failed shellcheck results include wiki links for each error code
Full logs are saved to a temp directory (path printed in output)
Overall result is printed at the end: Overall: PASS or Overall: FAIL

Important Notes

The script discovers all .sh files recursively, excluding .git, node_modules, and vendor
CHANGED_FILES scopes checks to only the listed .sh files that exist
SHELLCHECK_SEVERITY controls the minimum severity level for shellcheck warnings
In CI (CI=true), MAX_LINES defaults to unlimited; locally it defaults to 40

More from this repository

same repository

cargo-agent

codesoda/x-agent

Run cargo-agent.sh — a lean Rust workflow runner that produces agent-friendly output with structured diagnostics. Use when: running Rust checks (fmt, clippy, check, test), verifying Rust code before committing, or when the user asks to run cargo checks, lint, format, or test a Rust project. Triggers on: cargo agent, run cargo checks, rust checks, cargo fmt clippy test, verify rust code.

2026-03-102

sql-agent

codesoda/x-agent

Run sql-agent.sh — a lean SQL linter that produces agent-friendly output with sqlfluff. Use when: running SQL checks, linting SQL files with sqlfluff, validating SQL style, or when the user asks to run sqlfluff, sql lint, sql format, or sql checks. Triggers on: sql agent, sqlfluff, sql lint, sql format, sql checks, validate sql.

2026-03-082

ansible-agent

codesoda/x-agent

Run ansible-agent.sh — a lean Ansible playbook linter and syntax checker that produces agent-friendly output. Use when: running Ansible checks, linting playbooks with ansible-lint, syntax-checking playbooks, or when the user asks to validate ansible, run ansible lint, or check ansible syntax. Triggers on: ansible agent, ansible lint, ansible checks, validate ansible, ansible syntax check.

2026-03-082

docker-agent

codesoda/x-agent

Run docker-agent.sh — a lean Dockerfile linter that produces agent-friendly output. Use when: running Dockerfile checks, linting Dockerfiles with hadolint, verifying Dockerfiles before committing, or when the user asks to run hadolint, dockerfile lint, or docker checks. Triggers on: docker agent, dockerfile lint, hadolint, docker checks, validate dockerfile.

2026-03-082

kube-agent

codesoda/x-agent

Run kube-agent.sh — a lean Kubernetes manifest validator that produces agent-friendly output. Use when: running Kubernetes manifest validation, kubeconform checks, kubeval checks, verifying K8s manifests before applying, or when the user asks to validate Kubernetes YAML, run k8s checks, or check manifests. Triggers on: kube agent, kubernetes validate, kubeconform, kubeval, k8s checks, validate manifests.

2026-03-082

helm-agent

codesoda/x-agent

Run helm-agent.sh — a lean Helm chart linter and template validator that produces agent-friendly output. Use when: running Helm chart checks, linting charts, validating templates, verifying Helm charts before committing, or when the user asks to run helm lint, helm template, or Helm chart checks. Triggers on: helm agent, helm lint, helm checks, helm template, validate helm chart.

2026-03-082

name	bash-agent
description	Run bash-agent.sh — a lean shell script validation runner that produces agent-friendly output. Use when: running shell/bash script checks (syntax, lint), verifying shell scripts before committing, or when the user asks to run shellcheck, bash checks, or validate shell scripts. Triggers on: bash agent, shell agent, shellcheck, bash lint, shell checks, verify shell scripts.
context	fork
allowed-tools	["Bash(scripts/bash-agent.sh)","Bash(RUN_=* scripts/bash-agent.sh)","Bash(SHELLCHECK_SEVERITY= scripts/bash-agent.sh)","Bash(MAX_LINES= scripts/bash-agent.sh)","Bash(KEEP_DIR= scripts/bash-agent.sh)","Bash(FAIL_FAST= scripts/bash-agent.sh)","Bash(CHANGED_FILES= scripts/bash-agent.sh*)"]

Bash Agent

Run the bash-agent.sh script for lean, structured shell script validation output designed for coding agents.

Script Location

scripts/bash-agent.sh

Usage

Run Full Suite (syntax + lint)

scripts/bash-agent.sh

Run Individual Steps

scripts/bash-agent.sh syntax    # bash -n syntax check only
scripts/bash-agent.sh lint      # shellcheck lint only
scripts/bash-agent.sh all       # full suite (default)

Environment Knobs

Variable	Default	Description
`RUN_SYNTAX`	`1`	Set to `0` to skip syntax check
`RUN_LINT`	`1`	Set to `0` to skip shellcheck lint
`SHELLCHECK_SEVERITY`	`warning`	shellcheck `--severity` level (`error`, `warning`, `info`, `style`)
`FAIL_FAST`	`0`	Set to `1` to stop after first failure (or use `--fail-fast`)
`CHANGED_FILES`	(empty)	Space-separated changed file paths; scopes checks to those `.sh` files
`MAX_LINES`	`40`	Max output lines printed per step (unlimited in CI)
`KEEP_DIR`	`0`	Set to `1` to keep temp log dir on success

Output Format

Each step prints a header (Step: syntax, Step: lint)
Results are PASS, FAIL, or SKIP
On failure, output is truncated to MAX_LINES
Failed shellcheck results include wiki links for each error code
Full logs are saved to a temp directory (path printed in output)
Overall result is printed at the end: Overall: PASS or Overall: FAIL

Important Notes

The script discovers all .sh files recursively, excluding .git, node_modules, and vendor
CHANGED_FILES scopes checks to only the listed .sh files that exist
SHELLCHECK_SEVERITY controls the minimum severity level for shellcheck warnings
In CI (CI=true), MAX_LINES defaults to unlimited; locally it defaults to 40