agentmods.dev

semver-plugin CLAUDE.md

Project instructions for a portable Semantic Versioning rule pack used by several coding-agent tools. They describe the files, hooks, validation commands, and rules that keep the pack consistent.

In plain words
What is it for?
Use them when modifying the rule pack, its hooks, translations, plugin files, skills, or validation and test commands.
Why use it?
They show which file is authoritative and how changes affect every supported agent setup. They also provide checks for version files, translations, plugins, skills, and installation scripts.

Instructions file

1 1mo ago A scan 1,173 tokens original MIT
View the source file LenoSeibert/semver-plugin
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add instructions/lenoseibert/semver-plugin/claude-md
Clone the repo
git clone --depth 1 https://github.com/LenoSeibert/semver-plugin
Per session 1,173 This file is loaded in full into every session.
When invoked 1,173 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.01173 $0.01173
Opus 5 $0.00587 $0.00587
Sonnet 5 $0.00235 $0.00235
Haiku 4.5 $0.00117 $0.00117

Measured 2d ago against content hash edb009d1914b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

semver-plugin CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

CLAUDE.md · 89 lines

How it starts

The opening of the file, as written. The whole thing — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

What this repository ships

A portable Semantic Versioning 2.0.0 rule pack for five agent harnesses (Codex, Claude Code, Gemini CLI, GitHub Copilot CLI, OpenCode/Cursor). The product is Markdown and JSON, not a library. There is no build and no test suite.

AGENTS.md is the single source of the rule text: hooks/semver-context.js injects it as session context, gemini-extension.json declares it as the context file, and OpenCode/Cursor users copy it directly. Editing it changes behavior in every harness at once.

Commands

  • make validate — runs validate-version, validate-translations, validate-plugin, validate-skills, check-diff.
  • make test — smoke-tests scripts/install-local.sh with stubbed harness CLIs and --dry-run, asserting on the snapshot it builds. No harness is touched.
  • make validate-plugin and make validate-skills shell out to scripts outside this repo under $HOME/.codex/skills/.system/. They fail without Codex installed; override with PLUGIN_CREATOR_ROOT / SKILL_CREATOR_ROOT.
  • After changing skills/ or hooks/: run make validate, then make install-local and open a fresh session in the target harness to confirm the context injection actually landed. Static validation alone does not prove a hook works.
  • make lint — markdownlint-cli2 (via npx, needs network) plus shellcheck on scripts/*.sh. Kept out of make validate so validation stays offline; CI runs both.
  • make remove-local uninstalls from every harness.

Versioning invariants

  • VERSION and four manifests must be byte-identical: .codex-plugin/plugin.json, .github/plugin/plugin.json, .claude-plugin/plugin.json, gemini-extension.json. scripts/check-version.js fails CI when they drift. This is the most common way to break the build.
  • VERSION never carries a leading v; git tags do (v0.2.6).
  • Pre-1.0 policy for this repo, also stated in README.md: a breaking change to a published contract (hook event/command shape, skill name or frontmatter, manifest keys, dropping a harness) bumps MINOR; everything else bumps PATCH.
  • Every merge into main requires a version bump, docs-only changes included.
  • The bump is its own commit: chore(release): bump version to X.Y.Z.
  • .github/workflows/release.yml tags and publishes on push to main. It fails on purpose when vX.Y.Z already points at another commit — bump again rather than moving a released tag.

Read the full file on GitHub · 89 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 89 lines · 1,173 tokens per session scan A edb009d1914b

Subscribe to this mod's changes

semver-plugin CLAUDE.md is an instructions file published in the GitHub repository LenoSeibert/semver-plugin (1 stars, last pushed 1mo ago), licensed MIT. It adds 1,173 tokens to every session, about $0.0059 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Git and workflowReleasesai-agentsclaude-codecodexcoding-agentsdeveloper-toolsgemini-cli
Related

Other instructions, from other repositories

coral CLAUDE.md

Instructions for cdknorow/coral, covering claude.md - coral go, mission, testing, go unit tests and legacy parity tools (historical reference).

cdknorow/coral · 1,898 tokens

pingfusi CLAUDE.md

Instructions for alex-durango/pingfusi: This repository is the released form of the pingfusi kit (published to npm as pingfusi). Releases are assembled and verified by the maintainers' release tooling, so commits here are release-granular and PRs cannot be merged directly — open an issue or PR and a maintainer ports…

alex-durango/pingfusi · 182 tokens

Circus-Chief CLAUDE.md

Instructions for ferrislucas/Circus-Chief, covering claude.md, project overview, common commands, development - start both frontend and backend and run all unit tests.

ferrislucas/Circus-Chief · 2,092 tokens

obsidian-llm-hub AGENTS.md

Instructions for takeshy/obsidian-llm-hub, a project described as: Obsidian plugin for AI chat, workflow automation, and semantic search — supports Gemini, OpenAI, OpenRouter, Grok, Ollama, and CLI backends.

takeshy/obsidian-llm-hub · 44 tokens

deva AGENTS.md

Instructions for thevibeworks/deva: Repo guidance for coding agents lives in CLAUDE.md (architecture, issue-based development workflow) and workflows/ (issue, commit, PR, release conventions). Read those before changing anything.

thevibeworks/deva · 92 tokens

cupcake CLAUDE.md

Instructions for eqtylab/cupcake, covering cupcake system overview, policy engine (wasm/rego) details, 1. routing vs. policy execution (critical), 2. opa rego v1 migration (critical) and 3. decisions & synthesis.

eqtylab/cupcake · 855 tokens