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 skills add CUHK-Business-School-AI-Hub/coding_agent_constitution --skill constitution-skillgit clone --depth 1 https://github.com/CUHK-Business-School-AI-Hub/coding_agent_constitutionWrote 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/skills/cuhk-business-school-ai-hub/coding_agent_constitution/constitution-skill)<a href="https://agentmods.dev/skills/cuhk-business-school-ai-hub/coding_agent_constitution/constitution-skill"><img src="https://agentmods.dev/badge/skills/cuhk-business-school-ai-hub/coding_agent_constitution/constitution-skill/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/cuhk-business-school-ai-hub/coding_agent_constitution/constitution-skill"><img src="https://agentmods.dev/badge/skills/cuhk-business-school-ai-hub/coding_agent_constitution/constitution-skill.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00181 | $0.03894 |
| Opus 5 | $0.00090 | $0.01947 |
| Sonnet 5 | $0.00036 | $0.00779 |
| Haiku 4.5 | $0.00018 | $0.00389 |
Grade A, and why
constitution-skill 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 10d 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 — 310 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Constitution Skill
Purpose
Turn ambiguous software intent into durable project files before implementation. Treat specs, architecture, contracts, and rules as reusable assets; treat task plans, implementation notes, and one-off agent prompts as consumables.
This skill follows the portable SKILL.md shape so it can be installed as a Codex skill, Claude Code skill, or Cursor Agent Skill. The governance assets it creates should also work when different agents rotate through the same repository.
Modes
Pick the mode that matches the project before doing anything else.
| Mode | When | Output Footprint | Reference |
|---|---|---|---|
| Standard | Greenfield or growing project, multiple agents, real users expected | Full file set (~9 files) | This document, sections below |
| Retrofit | Existing codebase with no governance, mixed conventions, legacy code | Incremental, seam-first | references/retrofit-mode.md |
| Minimal | Solo, weekend, throwaway, no production traffic | 1-3 files | references/minimal-mode.md |
Default to Standard. Drop to Minimal only when at least three of these are true: single contributor, lifespan under 3 months, one coding agent, no external API, no auth, no persistent user data. Switch to Retrofit when the repo already has substantial code without governance assets.
Operating Model
Use this role split:
- Human owns product intent, boundaries, architecture decisions, risk approval, and merge decisions.
- Codex implements bounded units, edits files, runs checks, and produces reviewable changes.
- Cursor reviews diffs, checks architecture compliance, catches hidden coupling, and helps with local refinements.
- Claude Code can implement or review bounded units when it has the same governance files and tool-specific adapter files.
Keep one main editor per work cycle. Let Codex perform the main modification, then hand off a clean review surface to Cursor and the human.
Workflow
- Discover existing context.
- Read durable assets if present:
SPEC.md,docs/SPEC.md,ARCH.md,docs/ARCH.md,RULES.md,docs/RULES.md,CONTRACTS/,docs/CONTRACTS/,AGENTS.md,CLAUDE.md,.cursor/rules/,.claude/rules/. - Identify disposable assets if present:
TASKS/,docs/TASKS/, implementation notes, migration checklists, experiment logs. - Preserve existing project conventions and avoid overwriting durable assets without first understanding them.
- Read durable assets if present:
What ships with it
60 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- agents/openai.yaml 272 B
- assets/contracts-examples/cli-contract.md 2.8 KB
- assets/contracts-examples/db-schema.sql 2.0 KB
- assets/contracts-examples/event-payload.md 2.9 KB
- assets/contracts-examples/file-format.md 1.9 KB
- assets/contracts-examples/json-schema-validation.json 1.5 KB
- assets/contracts-examples/openapi-rest.yaml 4.6 KB
- assets/contracts-examples/README.md 923 B
- assets/examples/feedback-inbox/.claude/rules/project-governance.md 1.1 KB
- assets/examples/feedback-inbox/.cursor/rules/project-governance.mdc 1.2 KB
- assets/examples/feedback-inbox/AGENTS.md 3.0 KB
- assets/examples/feedback-inbox/CLAUDE.md 605 B
- assets/examples/feedback-inbox/docs/ARCH.md 5.8 KB
- assets/examples/feedback-inbox/docs/CONTRACTS/cli.md 2.2 KB
- assets/examples/feedback-inbox/docs/CONTRACTS/db-schema.sql 2.8 KB
- assets/examples/feedback-inbox/docs/CONTRACTS/events.md 3.3 KB
- assets/examples/feedback-inbox/docs/CONTRACTS/feedback-api.openapi.yaml 7.0 KB
- assets/examples/feedback-inbox/docs/CONTRACTS/file-format.md 1.6 KB
- assets/examples/feedback-inbox/docs/CONTRACTS/README.md 1.7 KB
- assets/examples/feedback-inbox/docs/DECISIONS/001-postgres-outbox-over-broker.md 1.5 KB
- assets/examples/feedback-inbox/docs/RULES.md 3.4 KB
- assets/examples/feedback-inbox/docs/SPEC.md 3.8 KB
- assets/examples/feedback-inbox/docs/TASKS/001-bootstrap-feedback-inbox.md 4.7 KB
- assets/examples/README.md 779 B
- assets/governance-templates/AGENTS.md 2.0 KB
- assets/governance-templates/ARCH.md 1.3 KB
- assets/governance-templates/claude-project-governance.md 862 B
- assets/governance-templates/CLAUDE.md 560 B
- assets/governance-templates/CONTRACTS_README.md 3.4 KB
- assets/governance-templates/cursor-project-governance.mdc 1023 B
- assets/governance-templates/DECISION.md 368 B
- assets/governance-templates/RULES.md 1.2 KB
- assets/governance-templates/SPEC.md 401 B
- assets/governance-templates/TASK.md 733 B
- assets/module-overlays/deterministic-workflow.md 1.7 KB
- assets/module-overlays/identity-access.md 2.5 KB
- assets/module-overlays/llm-boundary.md 2.2 KB
- assets/module-overlays/transactional-record-system.md 1.4 KB
- assets/templates/conversational-assistant.md 3.4 KB
- assets/templates/linear-workflow.md 2.9 KB
- assets/templates/record-crud.md 2.9 KB
- references/anti-patterns.md 9.8 KB
- references/bootstrap-question-bank.md 3.9 KB
- references/cross-agent-compatibility.md 4.0 KB
- references/governance-asset-guide.md 3.5 KB
- references/governance-evolution.md 5.5 KB
- references/governance-review-rubrics.md 3.9 KB
- references/minimal-mode.md 4.3 KB
- references/module-deterministic-workflow.md 3.0 KB
- references/module-identity-access.md 4.0 KB
- references/module-llm-boundary.md 4.2 KB
- references/product-pattern-routing.md 6.5 KB
- references/profile-transactional-record-system.md 4.4 KB
- references/recipe-local-python-sqlite.md 7.4 KB
- references/recipe-typescript-web-postgres.md 3.5 KB
- references/retrofit-mode.md 5.1 KB
- references/rookie-onboarding_CN.md 8.8 KB
- references/rookie-onboarding_HK.md 8.8 KB
- references/rookie-onboarding.md 9.0 KB
- references/task-review-contract.md 3.3 KB
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.
- 10d ago First seen · 310 lines · 181 tokens per session scan A 2dddbcbfee7b
constitution-skill is a skill published in the GitHub repository CUHK-Business-School-AI-Hub/coding_agent_constitution (4 stars, last pushed 2mo ago), licensed MIT. It adds 181 tokens to every session and 3,894 once invoked, about $0.0009 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.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…