agentmods.dev

agent-context-kit CLAUDE.md

A CLAUDE.md instruction file for AgentContextKit, a project that provides context-management tools for coding agents. It defines the project workflow, required checks, task documentation process, and repository rules.

In plain words
What is it for?
Use it when working on AgentContextKit to run the required version, doctor, scan, and task commands; follow task-first development; and check the repository before committing.
Why use it?
It tells agents how to use the project’s own command-line tool and prevents invalid validation steps or accidental commits of generated files.

Instructions file

14 4d ago A scan 935 tokens original MIT
View the source file Cynrath/agent-context-kit
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/cynrath/agent-context-kit/claude-md
Clone the repo
git clone --depth 1 https://github.com/Cynrath/agent-context-kit
Per session 935 This file is loaded in full into every session.
When invoked 935 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 $0.00935 $0.00935
Opus 5 $0.00467 $0.00467
Sonnet 5 $0.00187 $0.00187
Haiku 4.5 $0.00093 $0.00093

Measured 3d ago against content hash 85de5e845878, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agent-context-kit 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 3d 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 · 61 lines

How it starts

The opening of the file, as written. The whole thing — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Claude Project Context

Use the same repository rules as AGENTS.md (authoritative for all agents).

Dogfood / ACKit-First

This repository IS AgentContextKit. Every agent session must dogfood the tool.

  • Before ANY task: run node dist/cli/index.js --version, node dist/cli/index.js doctor, node dist/cli/index.js scan --ci.
  • Create new task docs with node dist/cli/index.js task "<title>" first, then fill/refine the generated Markdown.
  • Do not create task docs manually unless the CLI fails. If it fails, record the exact failure.
  • All validation uses the repository-built CLI: node dist/cli/index.js <command>. The old global .NET ackit tool and any dotnet run invocation are not valid validation tools on this branch.
  • Run node dist/cli/index.js doctor and node dist/cli/index.js scan --ci before every final commit.
  • Never commit generated .ackit/, reports, SARIF, prompt packs, context exports, package artifacts, packed tarballs, or temp outputs.
  • Preserve the task-first workflow, legacy-release immutability, and no-network/default safety rules.

Workflow

  • Task-first: every implementation change starts from docs/tasks/active/.
  • Continuous progress: do not stop between documented tasks; proceed through them in order.

Commit Completeness Hard Rule

  • Before any push, run git status and confirm the working tree is clean.
  • Never leave a newly created .md task file, plan, queue row, or test file uncommitted. New files must be added and committed in the same logical commit that creates them, or in an immediately following commit, before any push.

Stack

  • TypeScript (strict ESM) + Node.js >= 22
  • pnpm (packageManager pinned), Vitest, Biome
  • GitHub Actions: .github/workflows/ci.yml

Repository Health

  • README: yes
  • LICENSE: yes
  • SECURITY: yes
  • Tests: yes
  • CI: yes
  • Agent instructions: yes

Release Status

  • Current line: AgentContextKit, package @cynrath/agent-context-kit, version 0.2.0 on master (consolidated v0.2.0 release, tag v0.2.0, npm latest → 0.2.0, GitHub Release v0.2.0); development on master and PR branches running the same CI gate. 0.1.1/0.1.0 remain as historical releases.
  • npm publish, tags, and GitHub Releases are user-authorized actions only (see Controlled-release governance in AGENTS.md). Automated publishing runs ONLY through the tag-triggered .github/workflows/release.yml after explicit user authorization; master pushes never publish.
  • Legacy .NET line is frozen and immutable: NuGet AgentContextKit 1.0.0-rc.1 (exact tag/release/assets/attestations at repository commit 258918b33c3d1359aac967604ee524e8b66ddf02) and earlier prereleases remain untouchable; do not republish, reuse, move, replace, or delete them. Historical failed recovery evidence remains preserved.

Read the full file on GitHub · 61 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. 3d ago First seen · 61 lines · 935 tokens per session scan A 85de5e845878

Subscribe to this mod's changes

agent-context-kit CLAUDE.md is an instructions file published in the GitHub repository Cynrath/agent-context-kit (14 stars, last pushed 4d ago), licensed MIT. It adds 935 tokens to every session, about $0.0047 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 2026-08-30.

agents-mdclicodexcoding-agentsdeveloper-toolsrepository-scanner
Related

Other instructions, from other repositories

repository-harness AGENTS.md

Instructions for hoangnb24/repository-harness, covering agent instructions and harness.

hoangnb24/repository-harness · 332 tokens

repository-harness CLAUDE.md

Instructions for hoangnb24/repository-harness, covering project rules and harness.

hoangnb24/repository-harness · 61 tokens

boring-is-all-you-need AGENTS.md

Instructions for Khalilzhang0825/boring-is-all-you-need, covering agents.md - boring is all you need codex guide, project intent, working rules, validation and release discipline.

Khalilzhang0825/boring-is-all-you-need · 422 tokens

AGENTS.md AGENTS.md

Instructions for Anbeeld/AGENTS.md, covering global instructions, priorities, boundaries, uncertainty and evidence.

Anbeeld/AGENTS.md · 1,964 tokens

vibe-coding-repository-standard AGENTS.md

Instructions for cz1993/vibe-coding-repository-standard, covering vcrs repository instructions, purpose, communication, change discipline and privacy and safety.

cz1993/vibe-coding-repository-standard · 646 tokens

AgentRules AGENTS.md

Instructions for LearnAIHubC/AgentRules, covering agents.md, 1. think before coding, 2. simplicity first, 3. preserve existing architecture and framework boundaries and 4. surgical changes.

LearnAIHubC/AgentRules · 1,280 tokens