ejryan89-skill-core
Skill2 starsChanged 18 days agoMITView on GitHub
Internal Python modules shared by other skills' scripts. Not intended for direct invocation.
In this repository30 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- audit-repositorySkill
- change-skillSkill
- evaluate-skillSkill
- implement-changeSkill
- review-repositorySkill
- runtime-canarySkill
- shepherd-pull-requestsSkill
- audit-repositorySkill
- change-skillSkill
- evaluate-skillSkill
- implement-changeSkill
- review-repositorySkill
- runtime-canarySkill
- shepherd-pull-requestsSkill
- audit-ai-configSkill
- audit-skillSkill
- curate-agent-memorySkill
- dotnet-dependenciesSkill
- dotnet-formatSkill
- ejryan89-code-review-coreSkill
- flag-review-findingSkill
- github-activity-reportSkill
- repo-cleanupSkill
- review-documentSkill
- review-insightsSkill
- review-prsSkill
- update-coding-agent-skillsSkill
- update-pr-dashboardSkill
The file
Files
Skill core
This is a non-selectable dependency. Its ${CLAUDE_SKILL_DIR}/scripts/ folder holds Python modules that other skills' scripts import, such as console.py, whose runner starts every entry point with UTF-8 output and ends an unhandled failure with a FAILED line, frontmatter.py, which reads a SKILL.md or agent definition's frontmatter, github_client.py, the GitHub CLI client every gh call goes through, git_client.py, the git client, both bounded and kept from prompting by bounded_process.py, skill_roots.py, the directories the deployer installs skills in and the lookup of a sibling skill's scripts that every dependent puts on sys.path, flat_text.py, which flattens free text onto one line, folder_walk.py, the one walk of a repository's folders for files by name, which enters no junction or symbolic link its caller does not follow and skips one stated set of folders and every nested checkout, diff_paths.py, which decodes the paths a git diff's file headers name, typed_fields.py, the reader of a JSON field's kind that the code-review pipeline's and the pull request tracker's validators and the GitHub activity report share, whole_number.py, the one reader of a whole number from text, which takes ASCII digits alone up to the digit limit json_reader.py also holds, nesting_depth.py, the one bound on how deep a reader descends, which json_reader.py and xml_reader.py hold and a recursive parser checks at each level, versions.py, the one reader of a version number, which compares versions by their numbers, python_reader.py, the one parse of Python source and of a Python literal, which refuses source nested past the parser's limits as it refuses a syntax error, path_keys.py, the one form a path is compared in, json_writer.py, the one writer of JSON text, whether bound for a file, a prompt, a hook's output, or a message, which refuses at the write a non-finite number, a lone surrogate, and what json_reader.py would refuse reading it back, and xml_reader.py, which parses untrusted XML into an element tree after refusing a DOCTYPE or ENTITY declaration. It has no workflow to run.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

