Fable Discipline
assafkip/kipi-system
Teaches staged, verifiable coding habits for multi-file tasks and ships a hook that blocks tests from touching live data.
Agent skill
AI SAFE2 Development Method
by CyberStrategyInstitute in CyberStrategyInstitute/ai-safe2-framework
Plans and runs material repository changes with the AI SAFE2 method: classify delivery shape and risk, isolate the work, collect test evidence and finish with a completion receipt.
Install AI SAFE2 Development Method
$ npx skills add CyberStrategyInstitute/ai-safe2-framework --skill ai-safe2-development-method -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install CyberStrategyInstitute/ai-safe2-framework ai-safe2-development-method --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/CyberStrategyInstitute/ai-safe2-framework.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/codex/ai-safe2-development-method .claude/skills/ai-safe2-development-method && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "ai-safe2-development-method" agent skill from https://github.com/CyberStrategyInstitute/ai-safe2-framework/tree/main/skills/codex/ai-safe2-development-method into .claude/skills/ai-safe2-development-method/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ai-safe2-development-method", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/CyberStrategyInstitute/ai-safe2-framework/tree/main/skills/codex/ai-safe2-development-methodType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add CyberStrategyInstitute/ai-safe2-framework --skill ai-safe2-development-method -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install CyberStrategyInstitute/ai-safe2-framework ai-safe2-development-method --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/CyberStrategyInstitute/ai-safe2-framework.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/codex/ai-safe2-development-method .agents/skills/ai-safe2-development-method && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "ai-safe2-development-method" agent skill from https://github.com/CyberStrategyInstitute/ai-safe2-framework/tree/main/skills/codex/ai-safe2-development-method into .agents/skills/ai-safe2-development-method/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ai-safe2-development-method", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add CyberStrategyInstitute/ai-safe2-framework --skill ai-safe2-development-method -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install CyberStrategyInstitute/ai-safe2-framework ai-safe2-development-method --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/CyberStrategyInstitute/ai-safe2-framework.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/codex/ai-safe2-development-method .cursor/skills/ai-safe2-development-method && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "ai-safe2-development-method" agent skill from https://github.com/CyberStrategyInstitute/ai-safe2-framework/tree/main/skills/codex/ai-safe2-development-method into .cursor/skills/ai-safe2-development-method/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ai-safe2-development-method", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/CyberStrategyInstitute/ai-safe2-framework.git --path skills/codex/ai-safe2-development-method--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add CyberStrategyInstitute/ai-safe2-framework --skill ai-safe2-development-method -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install CyberStrategyInstitute/ai-safe2-framework ai-safe2-development-method --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/CyberStrategyInstitute/ai-safe2-framework.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/codex/ai-safe2-development-method .gemini/skills/ai-safe2-development-method && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "ai-safe2-development-method" agent skill from https://github.com/CyberStrategyInstitute/ai-safe2-framework/tree/main/skills/codex/ai-safe2-development-method into .gemini/skills/ai-safe2-development-method/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ai-safe2-development-method", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install CyberStrategyInstitute/ai-safe2-framework ai-safe2-development-methodInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add CyberStrategyInstitute/ai-safe2-framework --skill ai-safe2-development-method -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/CyberStrategyInstitute/ai-safe2-framework.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/codex/ai-safe2-development-method .github/skills/ai-safe2-development-method && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "ai-safe2-development-method" agent skill from https://github.com/CyberStrategyInstitute/ai-safe2-framework/tree/main/skills/codex/ai-safe2-development-method into .github/skills/ai-safe2-development-method/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ai-safe2-development-method", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add CyberStrategyInstitute/ai-safe2-framework --skill ai-safe2-development-method -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install CyberStrategyInstitute/ai-safe2-framework ai-safe2-development-method --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/CyberStrategyInstitute/ai-safe2-framework.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/codex/ai-safe2-development-method .opencode/skills/ai-safe2-development-method && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "ai-safe2-development-method" agent skill from https://github.com/CyberStrategyInstitute/ai-safe2-framework/tree/main/skills/codex/ai-safe2-development-method into .opencode/skills/ai-safe2-development-method/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ai-safe2-development-method", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
ai-safe2-development-method- GitHub stars
- 146
- Token cost
- ~833 tokens
- SKILL.md length
- 362 words
- Files
- 5 (incl. references)
- Skills in repo
- 3
- Repo updated
- First seen
- Licence
- Custom licence
At a glance
Plans and runs material repository changes with the AI SAFE2 method: classify delivery shape and risk, isolate the work, collect test evidence and finish with a completion receipt.
- Works in 10 steps: Read the nearest repository instructions… → Classify delivery shape independently… → Create a… → …
- Planning a risky repository change that needs a recorded design depth
- SKILL.md covers Workflow, Decision Rules and References
- Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
What it does
This skill turns software-development discipline into decision evidence, keeping policy, implementation, verification, semantic review and human authority separate. The agent reads repository instructions and policy, classifies the delivery shape as spike, bounded or architectural independently of risk, writes a plan source and runs safe2 dev plan to produce a plan. A review_required plan pauses only for the named missing decision, and an invalid plan is never worked around.
Repository-changing work is isolated and unrelated user changes are preserved. The evidence mode depends on the work: observe a failing then passing behavior for code, characterize ambiguous existing behavior first, establish the invalid and valid boundary for contracts and configuration, use render or interaction checks for documents and UI, and record hypothesis, result and discard decision for a spike. A failing check is localized before any fix, and reviews follow the cadence in the plan, with model reviewers treated as attributed evidence and a provider failure reported as unavailable rather than passing.
Before claiming completion the agent runs fresh verification on the final revision and produces a receipt with safe2 dev receipt. Architectural work and high or critical risk require explicit human design approval, and critical risk needs a threat model. It is not for read-only explanations or trivial edits, and it gives no authority to merge, release, deploy, accept risk or change policy.
When your agent uses it
- Planning a risky repository change that needs a recorded design depth
- Producing a completion receipt with fresh verification evidence
- Deciding whether a task is a spike, a bounded change or architectural work
- Gathering before-and-after evidence for a change
Example prompts
- “Plan the payment-webhook refactor with the SAFE2 method and tell me which delivery shape it is.”
- “Run the SAFE2 plan for this change and produce the completion receipt once the tests pass.”
- “Classify this auth change by delivery shape and risk before we write any code.”
Requirements
- The safe2 command line tool
Workflow steps
10 steps, taken from the first numbered list in SKILL.md.
- Read the nearest repository instructions and current policy before acting.
- Classify delivery shape independently from risk
- Create a safe2.development-plan-source.v1 source and run
- If the plan is review_required, pause only for the named missing decision.
- Isolate repository-changing work. Preserve unrelated user changes.
- Use the selected evidence mode
- When a check fails, localize the cause before proposing a fix. Do not stack
- Review at the cadence in the plan. Treat PR-Agent, Greptile, and other model
- Capture comparable before/after evidence when required. If no safe baseline
- Before any completion claim, run fresh verification against the final revision
What it can do on your machine
Read from SKILL.md and the folder at commit 9f92160. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
No scripts in the folder and no shell commands in SKILL.md.
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
AI SAFE2 Development Method loads about 833 tokens when it runs, and up to ~1.6k if it reads all its reference files. Until then it costs about 120 tokens; SKILL.md has 362 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
Its licence (Custom licence) doesn't allow us to republish the file, so here is its outline and opening line. It has 362 words (~833 tokens).
“Convert software-development discipline into decision evidence. Keep policy, implementation, verification, semantic review, and human authority separate.”
- name
- ai-safe2-development-method
Files
SKILL.md and 4 other files (references) in skills/codex/ai-safe2-development-method of CyberStrategyInstitute/ai-safe2-framework.
- SKILL.md
- SKILL-CARD.md
- agents/openai.yaml
- references/evaluation.md
- references/method.md
Open the folder on GitHubat commit 9f92160
Compare with similar skills
AI SAFE2 Development Method next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| AI SAFE2 Development Method this skillCyberStrategyInstitute/ai-safe2-framework | 146 | — | ~833 | Automated safety check: Pass | Custom licence | |
| Fable Disciplineassafkip/kipi-system | 112 | — | ~2.9k | Automated safety check: Pass | MIT | |
| Bulletproof Workflowartemiimillier/bulletproof | 153 | — | ~3.5k | Automated safety check: Pass | MIT | |
| TiDB Verification Profilespingcap/tidb | 41k | — | ~496 | Automated safety check: Pass | Apache-2.0 | |
| NIC Task Planningnginx/kubernetes-ingress | 5.1k | — | ~1.7k | Automated safety check: Pass | Apache-2.0 | |
| Feature Development WorkflowQwenLM/qwen-code | 28k | — | ~1.2k | Automated safety check: Pass | Apache-2.0 |
Similar skills
Bulletproof Workflow
artemiimillier/bulletproof
Applies a 12-stage verified workflow, from research to deploy, to non-trivial coding tasks, scaled to lightweight, standard or full mode by task size.
TiDB Verification Profiles
pingcap/tidb
Chooses how much validation a TiDB change needs: scoped checks while iterating, required checks at delivery, and expensive runs only when explicitly needed.
NIC Task Planning
nginx/kubernetes-ingress
Plans a change to the NGINX Ingress Controller before any code: acceptance criteria, security impact, affected layers, invariants, test surface and an ordered file list.
Feature Development Workflow
QwenLM/qwen-code
Phased workflow for non-trivial qwen-code features: investigate, design doc, E2E test plan, baseline dry-run, implement, verify, self-audit, review and iterate.
Engineering Plan Review
garrytan/gstack
Reviews an execution plan or design doc before coding, covering architecture, data flow, edge cases, test coverage and performance, one issue at a time.
More from CyberStrategyInstitute/ai-safe2-framework
AI SAFE2 Secure Build Copilot
CyberStrategyInstitute/ai-safe2-framework
Applies the AI SAFE2 framework to security reviews, code reviews and compliance mapping for AI agents, RAG pipelines and MCP servers.
AI SAFE2 Secure Build Copilot
CyberStrategyInstitute/ai-safe2-framework
Applies the AI SAFE2 v3.1 governance framework to designing, building, auditing and testing AI agents, RAG pipelines, MCP and tool integrations and AI infrastructure.
Related
Categories
Questions about AI SAFE2 Development Method
What does AI SAFE2 Development Method do?
Plans and runs material repository changes with the AI SAFE2 method: classify delivery shape and risk, isolate the work, collect test evidence and finish with a completion receipt. This skill turns software-development discipline into decision evidence, keeping policy, implementation, verification, semantic review and human authority separate. The agent reads repository instructions and policy, classifies the delivery shape as spike, bounded or architectural independently of risk, writes a plan source and runs safe2 dev plan to produce a plan.
When should I use AI SAFE2 Development Method?
AI SAFE2 Development Method fits situations like: planning a risky repository change that needs a recorded design depth; producing a completion receipt with fresh verification evidence; deciding whether a task is a spike, a bounded change or architectural work; gathering before-and-after evidence for a change.
How do I install AI SAFE2 Development Method in Claude Code?
Run `npx skills add CyberStrategyInstitute/ai-safe2-framework --skill ai-safe2-development-method -a claude-code`. Or copy the skill folder (skills/codex/ai-safe2-development-method in CyberStrategyInstitute/ai-safe2-framework) into .claude/skills/ai-safe2-development-method in your project. Claude Code loads it when a task matches its description.
How do I install AI SAFE2 Development Method in Codex?
Run `npx skills add CyberStrategyInstitute/ai-safe2-framework --skill ai-safe2-development-method -a codex`. Or copy the skill folder (skills/codex/ai-safe2-development-method in CyberStrategyInstitute/ai-safe2-framework) into .agents/skills/ai-safe2-development-method in your project. Codex loads it when a task matches its description.
Can I use AI SAFE2 Development Method in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add CyberStrategyInstitute/ai-safe2-framework --skill ai-safe2-development-method -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/ai-safe2-development-method, .gemini/skills/ai-safe2-development-method, .github/skills/ai-safe2-development-method and .opencode/skills/ai-safe2-development-method in your project.
What does AI SAFE2 Development Method need to run?
SKILL.md names no scripts, command-line tools or credentials: AI SAFE2 Development Method is instructions for the agent only. Our summary lists: The safe2 command line tool.
Does AI SAFE2 Development Method access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is AI SAFE2 Development Method safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does AI SAFE2 Development Method use?
AI SAFE2 Development Method has a licence file (the repository's licence) that doesn't match a standard licence. Read it on GitHub before reusing the skill.
How many tokens does AI SAFE2 Development Method use?
About 833 tokens (SKILL.md is roughly 3.3k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 761 tokens, read only when the agent opens those files.
What are the alternatives to AI SAFE2 Development Method?
Skills that share tags, products or a category with AI SAFE2 Development Method: Fable Discipline (assafkip/kipi-system, 112 stars), Bulletproof Workflow (artemiimillier/bulletproof, 153 stars), TiDB Verification Profiles (pingcap/tidb, 41k stars) and NIC Task Planning (nginx/kubernetes-ingress, 5.1k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains AI SAFE2 Development Method?
CyberStrategyInstitute (a GitHub organization) maintains it in CyberStrategyInstitute/ai-safe2-framework, which has 146 GitHub stars. The repository holds 3 skills in this directory. The repository was last updated on October 9, 2026.
Source: CyberStrategyInstitute/ai-safe2-framework on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.