milanhorvatovic

@milanhorvatovic source repo

23 skills
0 followers
9 hours ago last updated

▌
Git Toolkit · milanhorvatovic bundle
Use when about to commit staged work, authoring or amending any commit, opening or updating a PR, or preparing a release — routine authoring intent, not only explicit asks. Governs change narration across the git + GitHub lifecycle: branch naming and worktree setup, commit authoring, history cleanup, PR descriptions and issue linking, CI failure triage, review-thread resolution, merge readiness and execution, release notes. Enforces conventional-commits (when in use), imperative mood, ≤72-char subjects, and trailer placement. Also fires on asks to write, validate, review, fix, sync, or clean up that text, and when commits feel inconsistent. Works on GitLab, Codeberg/Forgejo, and Bitbucket Cloud. Never auto-publishes or auto-adds trailers. Read-only inspection (status, log, diff) stays out.

0
installs
▌
Docs Steward · milanhorvatovic bundle
Use when markdown docs need checking or fixing, not writing: the skill audits and formats a repo's markdown by orchestrating the formatter its config declares (markdownlint-cli2 / markdownlint / prettier / mdformat / dprint / remark) plus a complementary lint pass, and runs a yamllint audit over frontmatter and fenced YAML blocks; bundled fallbacks fill what the repo leaves undeclared. Read-only until fixes are approved; emits NDJSON findings the invoking agent renders and offers to fix. Triggers on "steward the docs", "audit docs", "check docs", "review markdown", "format docs", "fix markdown", "audit frontmatter", "lint frontmatter", "recommend doc tools", pre-release sweeps, or /docs-steward. Never writes new prose, adds no rules beyond its bundled fallbacks, never auto-installs tools.

0
installs
▌
Behavior Coach · milanhorvatovic bundle
Distills the observable working behavior of a stronger source model into a portable, spec-compliant skill any target model can load — process transfers (decomposition, verification discipline, next-action policy, reporting style); raw capability does not. Six-stage pipeline: scope, capture via probe battery, baseline the bare target, extract deltas, author under portability rules, pressure-test RED/GREEN/REFACTOR. Prompt-level transfer only — never fine-tuning, training-data generation, or system-prompt extraction. Triggers on distill / clone / capture / preserve a model's behavior or reasoning style, make model B work like model A, write a skill teaching a smaller model how this one works, or porting a deprecated or repriced model's process before access changes, or /behavior-coach.

0
installs
▌
Toolchain Doctor · milanhorvatovic bundle
Examines a repository's per-language tooling and prescribes what is missing. scan inventories the lint, format, type-check, and version-pin declarations and whether CI runs them; audit grades the distance to each language's tooling floor, advisory only; scaffold proposes minimal pinned configs and CI steps, one confirmation each. Covers python (ruff, mypy or pyright), typescript (strict tsc, biome or eslint plus prettier), rust (cargo fmt, clippy at deny-warnings), and bash (shellcheck, shfmt). Never installs anything, never writes unconfirmed. Triggers on "set up linting for this repo", "what tooling is this project missing", "is our formatter setup right", "add a type checker", "why doesn't CI run the linter", "pin our tool versions", or /toolchain-doctor.

0
installs
▌
Coding Principles · milanhorvatovic bundle
Implementation discipline for any coding task — writing, modifying, fixing, refactoring, or reviewing code. Carries 16 universal mantras (tiers: goals, design, pruning) and 21 numbered principles with must/should/could severity, with on-demand language capabilities (bash, python, typescript, rust), a cross-language comments capability, and a review workflow. Triggers when the user asks to write, implement, add, fix, refactor, clean up, or review code, on planning docs (plans, ADRs, work specs) that mention comments or docstrings, or via /coding-principles. Skip for other docs-only, config-only, pure-data, or ops/infra tasks.

0
installs
▌
Oss Repository Conventions · milanhorvatovic bundle
Stewards an open-source repo toward and along a top-notch standard. Triggers on audit / set up / harden / level up an OSS repo, "what is this repo missing", "score its health", "add a SECURITY policy / license / CONTRIBUTING", "set up auto-merge", "auto-approve bot PRs", "make CI composable", first-touch onboarding; invoked on /oss-repository-conventions. Covers licensing; security and governance; contribution, conduct, and community health; repo infrastructure and dev setup; code style and testing; CI automation, dependency supply chain, and PR autonomy; releases and documentation. Modes: scan (what's declared), audit (gaps scored by severity), scaffold (drafts missing files, one confirmation each). Reports and proposes; not for commit/PR/branch/release-notes prose (change-narration).

0
installs
▌
Skill System Foundry · milanhorvatovic bundle
Designs and evolves AI-agnostic skill systems. Triggers on skill/capability creation, role definition, or router migration; use when auditing consistency or improving token efficiency.

0
installs
▌
Review · milanhorvatovic
Guides human PR review and contributor self-review for the Skill System Foundry repository. Triggers when preparing a pull request for others to review, doing a self-review before requesting review, reviewing someone else's PR on GitHub, deciding whether to approve or request changes, evaluating whether automated review findings are worth addressing, or understanding what to check before merging. Also triggers when asked about the review process, review checklists, or how to structure review feedback for contributors. This skill provides process guidance — for running automated checks locally, use the local-code-review skill instead.

0
installs
▌
Critique · milanhorvatovic
Provides constructive criticism of proposed plans and implemented solutions in the Skill System Foundry repository. Challenges approach decisions, identifies weaknesses, trade-offs, and blind spots before changes reach automated checks or PR review. Triggers when asked to critique a solution, challenge an approach, find weaknesses in an implementation, review a plan before implementation, play devil's advocate, or provide constructive feedback on design decisions. Also triggers on phrases like "what could go wrong," "what am I missing," "is this the right approach," "poke holes in this," or "give me honest feedback." Use this skill after implementation and before running local-code-review to catch design-level issues that automated checks cannot detect.

0
installs
▌
Git Release · milanhorvatovic
Guides the release lifecycle for the Skill System Foundry repository — version bumping, changelog preparation, GitHub release creation, and post-release verification. Triggers when asked to prepare a release, bump a version, create a changelog, tag a release, publish a GitHub release, verify a release artifact, or update the version in SKILL.md frontmatter. Also triggers on questions about semver conventions, release workflows, the release.yaml GitHub Action, or distributing skill bundles. Use this skill whenever version numbers, releases, tags, or distribution are mentioned.

0
installs
▌
YAML Config · milanhorvatovic
Governs the configuration.yaml file that serves as the single source of truth for all validation rules in the Skill System Foundry. Triggers when adding, modifying, or reviewing validation rules, limits, patterns, or reserved words. Also triggers when working with constants.py, yaml_parser.py, or any code that reads from configuration.yaml. Use this skill when asked to add a new validation check, change a limit or threshold, update reserved word lists, add SPDX license identifiers, modify regex patterns, or troubleshoot why a validation rule is not working as expected. Activates on mentions of configuration, validation rules, constants, thresholds, or pattern definitions.

0
installs
▌
Markdown Docs · milanhorvatovic
Enforces documentation quality and structural consistency across Markdown files in the Skill System Foundry repository. Triggers on creating, editing, reviewing, or auditing any .md file — including SKILL.md entry points, reference documents, README files, capability.md files, role definitions, and CONTRIBUTING.md. Also triggers when asked to check documentation consistency, fix cross-references, improve descriptions, enforce progressive disclosure, review frontmatter formatting, or align terminology across files. Use this skill whenever Markdown content is being created or modified, even if the user does not explicitly mention "documentation."

0
installs
▌
Shell Scripts · milanhorvatovic
Enforces quality, safety, and consistency for shell scripts in the Skill System Foundry .github/scripts/ directory. Triggers when creating, editing, reviewing, or debugging any .sh file under .github/scripts/, modifying GitHub Actions workflows that call shell scripts, or working with CI automation scripts. Also triggers when asked to add a new CI helper script, fix shellcheck violations, handle GitHub Actions environment variables, or write new .sh scripts for CI. Use this skill for any shell scripting work in the repository.

0
installs
▌
Hello Router · milanhorvatovic bundle
Greets a recipient through one of two registered tones — formal or casual — by dispatching to a dedicated capability. Activates when the conversation asks for a tone-specific welcome or a switch between formal and casual greetings; use when comparing the two styles. Demonstrates the router pattern in the Skill System Foundry — a thin SKILL.md entry point routing to capability files, with allowed-tools declared in frontmatter so capability shell fences pass validation.

0
installs
▌
Github Actions · milanhorvatovic
Guides the creation, modification, and review of GitHub Actions workflows in the Skill System Foundry repository. Covers workflow YAML syntax, trigger configuration, permission models, SHA-pinned actions, matrix strategies, artifact passing, and concurrency control. Triggers when asked to create or modify a workflow, add a CI job, configure triggers or permissions, pin an action to a SHA, troubleshoot a failing workflow, or review a workflow change. Also triggers on phrases like "add a workflow," "fix the CI," "pin this action," "workflow permissions," or "why is this action failing." Use this skill for any work involving .github/workflows/ YAML files.

0
installs
▌
Python Scripts · milanhorvatovic
Enforces code quality, modularity, and test coverage for Python scripts in the Skill System Foundry repository. Triggers when creating, editing, reviewing, refactoring, or debugging any Python file under scripts/ or scripts/lib/, writing or updating tests under tests/, adding new validation logic, modifying shared library modules, creating new entry points, or working with the configuration.yaml validation rules. Also triggers when asked to improve test coverage, enforce type hints, fix error handling, extract shared logic, or ensure stdlib-only compliance. Use this skill for any Python work in the repository, even if the user does not explicitly mention code quality.

0
installs
▌
Hello Greeter · milanhorvatovic
Greets a single recipient with a friendly welcome message rendered in a formal or casual tone. Activates when the conversation asks to say hello or welcome someone; use when a minimal standalone skill is needed. Demonstrates the smallest valid standalone skill in the Skill System Foundry — required name and description frontmatter plus an optional metadata block — and how its layout passes validation.

0
installs
▌
Skill Authoring · milanhorvatovic
Guides the creation, evolution, and validation of skills within the Skill System Foundry ecosystem. Triggers when asked to create a new skill, scaffold a skill from a template, add capabilities to an existing skill, migrate a standalone skill to a router pattern, define a role, write or improve a skill description, set up a manifest, validate a skill against the Agent Skills specification, audit a skill system, bundle a skill for distribution, or decide between standalone and router architectures. Also triggers on questions about skill structure, progressive disclosure, token economy, or cross-platform deployment.

0
installs
▌
Solution Design · milanhorvatovic
Guides the design and validation of a solution before implementation begins. Breaks a task into steps, identifies affected files, evaluates trade-offs, and validates the plan is sound. Triggers when asked to plan a solution, design an approach, think through a change before coding, break down a task, or validate a proposed plan. Also triggers on phrases like "how should we approach this," "plan this change," "what's the best way to implement this," "break this down," or "let me think through this first." Use this skill as the first step in the development workflow — before implementation, critique, and local code review.

0
installs
▌
Local Code Review · milanhorvatovic
Runs a local code review workflow for the Skill System Foundry repository — executes tests, coverage checks, skill validation, and shellcheck, then analyzes the diff for pattern violations and produces a structured verdict. Triggers when asked to check code locally, run a local review, verify changes pass all checks, or catch issues before CI. Also triggers on phrases like "check this before I push," "does this look right," "run the checks," or "review the current branch locally." Use this skill to catch issues locally and avoid spending CI tokens and time on problems that can be detected on the development machine. For a deep review replicating the CI code review pipeline with confidence scoring and structured findings, use the local-ci-code-review skill instead. For design-level criticism and constructive feedback, use the critique skill instead. For human PR review process guidance, use the review skill instead.

0
installs
▌
Commit Conventions · milanhorvatovic
Enforces commit message format and conventions for the Skill System Foundry repository. Triggers when composing a commit message, reviewing commit history, squashing commits, or deciding how to describe a change. Also triggers on phrases like "write a commit message," "what should the commit say," "commit message format," or "how to reference an issue in a commit." Use this skill whenever a commit message needs to be written or reviewed.

0
installs
▌
Validate Skill Spec · milanhorvatovic
Validates a skill directory against the Agent Skills specification (agentskills.io/specification). Checks file reference consistency, frontmatter compliance, progressive disclosure, and structural correctness. Triggers when asked to validate a skill against the spec, check skill references, verify spec compliance, audit a skill's structure, or confirm a skill is ready for distribution. Also triggers on phrases like "is this skill valid," "check spec compliance," "validate file references," "verify the skill structure," or "does this skill follow the spec." Use this skill to catch structural issues, broken references, and spec violations in any skill directory.

0
installs
▌
Local CI Code Review · milanhorvatovic
Runs a deep structured code review locally, applying the same methodology used by the codex-ai-code-review-action CI pipeline — priority levels, confidence scoring, data flow tracing, and structured findings — without requiring GitHub Actions or an external API. Triggers when asked to run the CI review locally, do a deep code review, review like the pipeline would, or check what CI would flag. Also triggers on phrases like "run the CI review on this," "what would the pipeline review find," "deep review this branch," or "simulate the automated review." For running automated checks (tests, coverage, shellcheck), use the local-code-review skill instead. For human PR review process guidance, use the review skill instead.

0
installs

23 published skills