Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    CLI Help Docs

    1

    Automated CLI documentation generator that harvests help trees and verifies examples against the actual binary.

    $5.99

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    CLI Help Docs

    CLI Help Docs

    Example session with this skill installed

    The README for our 'deploy-tool' is totally out of date after the v2.0 release. Audit the current help text and update the usage section with verified examples.

    • Read your context and instructions
    • Compiled the cli help docs

    I have harvested the help tree for deploy-tool v2.0.1. I found 4 new subcommands (provision, rollback, logs, status) and 3 renamed flags that were not present in your current README.

    CLI Reference Audit

    CommandStatusIssues Found
    deploy-tool provisionMissingSubcommand not found in README
    deploy-tool logsOutdatedFlag --tail in README is now --lines in binary
    deploy-tool rollbackMissingSubcommand not found in README
    deploy-tool initBrokenExample flag --env-file now requires a path

    Updated Usage Section

    <!-- cli-reference: generated from deploy-tool v2.0.1 on 2024-05-20 -->

    deploy-tool provision

    Initialize infrastructure for a specific environment.

    Synopsis
    deploy-tool provision <env> [options]

    Flags

    • -r, --region <string>: Target cloud region. (Default: "us-east-1")
    • -c, --count <int>: Number of instances to spin up. (Default: 3)
    • --dry-run: Validate config without applying changes.

    Examples

    deploy-tool provision staging --region us-west-2 --dry-run # safe
    deploy-tool provision production --count 10 # side effects
    
    <!-- /cli-reference -->

    Next steps

    1. Review the generated provision block for accuracy against your intended workflow.
    2. Run scripts/check-doc.sh to confirm the production example matches your updated schema.
    3. Replace the old "Usage" heading in README.md with this verified content.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Documentation that relies on memory or source code comments eventually rots. A flag is renamed, a default changes, or a subcommand is added, and the README quickly becomes a liability. This skill ensures your CLI documentation is a precise reflection of the actual binary, not a guess.

    What it does

    • Help tree harvesting scans the entire CLI command structure to capture flags, arguments, and defaults directly from the binary.
    • Drift detection identifies gaps between the current documentation and the actual help output, flagging missing subcommands or renamed flags.
    • Example verification runs "safe" examples against the binary to ensure they exit 0 and statically validates "side effect" examples against harvested data.
    • Multi-format generation produces README usage sections, standalone reference pages, or POSIX-compliant man pages.
    • Automated stamping wraps generated content in HTML comments containing version and timestamp data for clear provenance.

    How it works

    1. Harvest the help tree using the provided script to generate a command manifest and individual help dumps.
    2. Select the output format based on the audience, such as a deep-dive reference or a high-level README overview.
    3. Verify every example in the document using a checker that executes safe code and audits flags against the harvest.

    Frameworks & tools

    This skill utilizes shell scripts for binary interrogation, Markdown for output, and standard CLI conventions. It requires a runnable binary of the tool being documented.

    Why this beats prompting it yourself

    Generic LLMs often hallucinate CLI flags or invent syntax based on common patterns that your specific tool might not follow. This skill ties every word to a physical help dump, ensuring that every default value and flag combination is verified against your actual implementation before it ships.

    Use cases

    • Auditing a mature CLI to find undocumented features or deprecated flags.
    • Generating a standard man page for a new internal tool distribution.
    • Syncing a project README after a major version bump with breaking changes.
    • Verifying that code examples in your documentation actually run and pass.

    Known limitations

    The skill cannot document hidden flags or internal commands unless explicitly instructed. It requires the CLI binary to be accessible and runnable in the environment.

    Requires a runnable binary. Cannot document hidden flags without explicit override. Static analysis is limited by the quality of the tool's --help output.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 1 month ago

    Permissions

    Terminal / Shell
    Listed1 month ago

    What's inside

    Frequently Asked Questions

    This skill eliminates 'documentation rot' by directly harvesting data from your CLI binary rather than relying on memory or source code, ensuring every flag, argument, and default is 100% accurate.

    The tool can generate POSIX-compliant man pages, comprehensive standalone reference guides, or concise README usage sections, all including automated timestamps and version stamps for provenance.

    The skill performs 'safe execution' on non-destructive examples to ensure an exit 0 status and uses static validation against the harvested help tree to audit examples that might have side effects.

    You will receive the core help tree harvesting scripts, the documentation generation engine, and the validation suite, along with a guide on integrating these into your CI/CD pipeline.

    The skill requires the target CLI binary to be runnable in the environment; it supports any CLI that follows standard help output conventions (e.g., --help) regardless of the language it was written in.

    New in Technical Documentation