CCS is a command-line manager that lets users switch among AI coding runtimes, provider accounts, and model profiles through one interface. It is for developers who use Claude Code or compatible CLIs and need to change accounts or backends without repeatedly editing configuration.
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.
npx agentmods add instructions/kaitranntt/ccs/claude-mdgit clone --depth 1 https://github.com/kaitranntt/ccsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/kaitranntt/ccs/claude-md)<a href="https://agentmods.dev/instructions/kaitranntt/ccs/claude-md"><img src="https://agentmods.dev/badge/instructions/kaitranntt/ccs/claude-md.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.01421 | $0.01421 |
| Opus 5 | $0.00711 | $0.00711 |
| Sonnet 5 | $0.00284 | $0.00284 |
| Haiku 4.5 | $0.00142 | $0.00142 |
Grade A, and why
ccs 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 5d 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.
How it starts
The opening of the file, as written. The whole thing — 145 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CCS CLI Agent Guide
Canonical agent instructions for /Users/kaitran/CloudPersonal/ccs/cli.
AGENTS.md must stay a symlink to this file.
Scope
CCS is a TypeScript/Bun CLI and dashboard for managing Claude Code, Codex, Factory Droid, CLIProxy, and compatible provider profiles.
Non-Negotiables
- Default branch is
dev. Feature/fix branches start fromdev; production hotfixes start frommainonly when explicitly needed. - Never touch the user's real
~/.ccs/or~/.claude/in tests. UsegetCcsDir()fromsrc/utils/config-manager.ts; it respectsCCS_HOME. - Do not commit directly to
devormain. - Do not manually bump versions or create release tags. Semantic-release owns versions, changelog, tags, npm publish, and GitHub releases.
- CLI terminal output must be ASCII only:
[OK],[!],[X],[i]. - Respect
NO_COLORand TTY-aware output.
Architecture
src/- TypeScript CLI/server source.lib/ccs,lib/ccs.ps1- bootstrap wrappers; no help text here.ui/src/- React dashboard.dist/anddist/ui/- build outputs.docs/- local development and architecture docs.- Docker support lives under
docker/and related commands.
Profile resolution priority:
- Built-in CLIProxy providers: Gemini, Codex, Antigravity.
- User-defined
config.cliproxyproviders. - Settings-based
config.profiles. - Account-based
profiles.jsonwith isolatedCLAUDE_CONFIG_DIR.
All env values written into settings must be strings.
Documentation Truth
Use the narrowest authoritative source:
- Source, tests, package scripts, and workflows define implemented behavior.
CLAUDE.mdandCONTRIBUTING.mddefine repository workflow.docs/README.mdmaps maintainer documentation and its owners.- The separate
kaitranntt/ccs-docsrepository and published site own user guides and CLI reference. - Generated artifacts and live runtime checks define what shipped or is currently running.
Do not copy inventories, line counts, locale lists, target lists, or command details when a stable source link is enough. Update the owning documentation when behavior, commands, setup, architecture, security posture, or maintainer workflow changes. Remove stale claims instead of preserving them as TODOs.
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.
- 5d ago First seen · 145 lines · 1,421 tokens per session scan A 5ff8d81aa0d3
ccs CLAUDE.md is an instructions file published in the GitHub repository kaitranntt/ccs (2,838 stars, last pushed 9d ago), licensed MIT. It adds 1,421 tokens to every session, about $0.0071 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.
Other instructions, from other repositories
CPA-Manager-Plus AGENTS.md
AGENTS.md instructions for seakee/CPA-Manager-Plus, covering repository rules, operation, layout, commands and architecture.
agentgateway copilot-instructions.md
Copilot instructions for agentgateway/agentgateway: Do not check for, speculate about, or report compilation errors during code review. Compilation diagnostics from review are frequently incorrect; rely on CI to detect and report compilation failures.
gmgn-skills CLAUDE.md
Claude Code instructions for GMGNAI/gmgn-skills, covering claude.md, critical rule — read this first, project overview, available skills and quick decision guide.
1flowbase AGENTS.md
AGENTS.md instructions for taichuy/1flowbase, covering 记忆, 用户偏好, 本项目相关skill在, 本项目 skills and 质量控制.
aisix CLAUDE.md
Claude Code instructions for api7/aisix, covering claude.md, 1. think before coding, 2. simplicity first, 3. surgical changes and 4. goal-driven execution.
core copilot-instructions.md
Copilot instructions for asdecided/core: These decisions are already accepted. Do not re-open or contradict them; ask the AsDecided MCP tools (getartifact, searchartifacts) for the full text before proposing a change that touches one.