Lets a Hermes agent turn a task it keeps repeating into a tested, reusable skill, using askit to lint, scan, test and score the result before sharing it. Use when the same kind of request comes up more than twice, or when the user asks the agent to save a workflow as a skill.
The English summary is being prepared. Showing the original description.
Run score Not tested
See how to installReview the instructions the agent will receive before you install.
A playbook for an agent that notices repeated work and captures it as a skill, with checks that do not depend on the agent's own opinion of its work.
Search first. Look through installed skills and memory for something that already covers the task. Extend an existing skill instead of writing a near copy.
Abstract. Remove details that belong to one run: names, paths, dates. Decide the inputs, the output and what should happen on bad input.
Write. Create ~/.hermes/skills/<name>/SKILL.md with frontmatter (name, description, license, metadata)
and short instructions. Add scripts/ when code is more reliable than prose.
Test. Add a cases.json file inside a tests folder with a happy path, an edge case and an error case, then run the checks:
askit lint ~/.hermes/skills/<name>
askit scan ~/.hermes/skills/<name>
askit test ~/.hermes/skills/<name>
askit score ~/.hermes/skills/<name>
Refine. If the score is below 0.8, read the failing checks, fix the skill and run them again.
Share. When the gate passes, run askit publish ~/.hermes/skills/<name> and tell the user where the package is.
Remember. Store one line in memory: skill name, score and what it is for, so step 1 finds it next time.
askit score.If askit is not installed, install it with pip install agentskill-toolkit and continue. If the checks cannot run,
say so and do not claim the skill is tested.
No reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Runs a list of HTTP checks against a base URL from a JSON file and reports status codes, response time and body assertions. Use when the user asks whether an API is up, wants a smoke test after a deploy, or needs to verify that endpoints return what the docs promise.
The English summary is being prepared. Showing the original description.
Turns git log output into a grouped markdown changelog using conventional commit prefixes such as feat, fix and docs. Use when the user prepares a release, asks for release notes, or wants a summary of what changed between two tags.
The English summary is being prepared. Showing the original description.
Writes clear git commit messages in the conventional commits style from a diff or a description of the change. Use when the user asks for a commit message, wants to tidy up a message before pushing, or has staged changes and no wording yet.
The English summary is being prepared. Showing the original description.
Profiles a CSV file column by column, reporting inferred type, missing values, unique counts and min, max and mean for numeric columns. Use when the user hands over a CSV and asks what is in it, whether the data is clean, or what to check before analysis.
The English summary is being prepared. Showing the original description.
Checks a Dockerfile for common problems such as unpinned base images, running as root, ADD instead of COPY, secrets in ENV or ARG and apt-get without cleanup. Use when the user asks to review, harden or slim down a Dockerfile.
The English summary is being prepared. Showing the original description.
Compares a .env file with its .env.example template and reports missing keys, undocumented extra keys, empty values and example files that contain real looking secrets. Use when the user asks why an app fails on startup with missing configuration, or wants to check environment files before committing.
The English summary is being prepared. Showing the original description.