Agent skill
init
This skill should be used when a user asks to initialize a repository for AI agents, create or update an AGENTS.md file, scaffold agent instructions for a codebase, or improve existing agent guidance. Triggers include phrases like 'create an AGENTS.md', 'set up agent instructions', 'initialize this repo for coding agents', 'update my AGENTS file', or 'analyze this repo and generate agent docs'.
Install this agent skill to your Project
npx add-skill https://github.com/srtab/daiv/tree/main/daiv/automation/agent/skills/init
SKILL.md
Analyze this repository and produce the full contents of AGENTS.md for future coding agents.
Goal (very important):
- Make this file as SHORT as possible while still capturing REPO-SPECIFIC, NON-OBVIOUS, HIGH-LEVERAGE guidance.
- Avoid redundancy: do not restate README/docs/rules unless you are adding a concise delta that changes agent behavior.
- Prefer pointing to existing docs by path (e.g., "See docs/xyz.md") rather than copying text.
Hard constraints:
- Target length: 250–500 words. Only exceed if there is a critical repo-specific reason.
- Only include commands that you can VERIFY exist in the repo (Makefile, scripts, pyproject, CI files, docs). Do not guess.
- Do not add generic engineering advice, generic style guides, or “be careful” warnings.
- Do not invent sections (e.g., “Tips”, “Support”, “Common tasks”) unless you found that exact content in-repo and it is essential.
- Do not include “ask for confirmation” directives; instead, express risk as plan-time checks.
If AGENTS.md already exists:
- Produce an updated version that reduces duplication and removes low-signal text.
- Keep any existing repo-specific invariants and verified commands, but rewrite for brevity and clarity.
Inputs to consult (only extract what is truly needed and not redundant):
- Existing
AGENTS.md(if present) .cursor/rules/,.cursorrules,CLAUDE.md,.github/copilot-instructions.md(if present)README.mdand docs (only for verified commands / invariants)- Build/test/lint definitions: Makefile, pyproject.toml, CI workflow files, scripts directories
Required output format:
- Produce ONE Markdown document: the full contents of
AGENTS.md. - Prefix EXACTLY:
```
AGENTS.md
This file provides guidance to agents when working with code in this repository. ```
See examples/agents-md-sample.md for a reference output demonstrating all four sections done well.
Sections (keep terse; omit any section if you have nothing verified to add):
- Commands (verified): fastest path for test + lint + single-test invocation
- Repo map (only what’s non-obvious): 5–10 bullets max; name directories only if it helps navigation
- Invariants / footguns: only rules that prevent wasted work or incorrect changes (e.g., state update patterns, “don’t edit X directly”)
- Where changes usually go (optional): only if there are non-obvious entry points that save time
Writing style:
- Bullets over paragraphs.
- No long examples. No duplicated explanations.
- If unsure, omit.
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
security-audit
This skill should be used when the user asks to review code for security vulnerabilities, audit a pull request or merge request for risks, check if code is safe, find injection flaws or hardcoded secrets, or assess the security posture of a feature or codebase area. Use this skill even when the user doesn't say "security audit" explicitly — trigger on phrases like "is this code safe?", "audit this PR", "check for vulnerabilities", "any security issues here?", "review my changes for risks", "are there hardcoded secrets?", "check for SQL injection", "security review", or "assess the risk in this diff". When in doubt, use this skill — it's better to over-trigger on security reviews than to miss one.
code-review
This skill should be used when a user asks for a code review, feedback on a PR or MR, diff assessment, or says things like 'can you review my changes', 'look at this diff', 'is this ready to merge', 'check my code', 'review this branch', 'what do you think of these changes', or 'LGTM check'. Covers correctness, tests, performance, security, and architecture feedback on pull/merge requests or raw diffs from any platform (GitHub, GitLab).
plan
This skill should be used when the user asks to explore a codebase and design implementation plans without making any changes. Trigger when users say 'plan how to implement X', 'design an approach for Y', 'explore the codebase before changing Z', 'create an implementation strategy', 'analyze how to refactor X', 'map out dependencies for Y', 'I want a plan before we start coding', 'plan to fix issue
skill-creator
Guide for creating effective skills that extend DAIV agent with specialized knowledge, workflows, or tool integrations. Use this skill when the user wants to create a new skill, update an existing skill, or get guidance on skill design patterns.
verl-rl-training
Provides guidance for training LLMs with reinforcement learning using verl (Volcano Engine RL). Use when implementing RLHF, GRPO, PPO, or other RL algorithms for LLM post-training at scale with flexible infrastructure backends.
openrlhf-training
High-performance RLHF framework with Ray+vLLM acceleration. Use for PPO, GRPO, RLOO, DPO training of large models (7B-70B+). Built on Ray, vLLM, ZeRO-3. 2× faster than DeepSpeedChat with distributed architecture and GPU resource sharing.
Didn't find tool you were looking for?