Role skill pack design
Skill markoblogo/abvx-agent-skills/skills/role-skill-pack-design
Design role-specific or workflow-specific agent skill packs without collapsing everything into one prompt. Use when creating assistant layers for teams, products, functions, or operating roles. Define base-vs-difference layers, pack boundaries, references, triggers, and rollout order.From its SKILL.md
npx -y skills add markoblogo/abvx-agent-skills --skill role-skill-pack-designAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 4 stars4 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its file declares
Copied from the file, not written here
The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.
SKILL.md
2.4 KB, 442 tokens by cl100k_base, as published. Nobody here has run it
Role Skill Pack Design
Design a reusable skill pack for role-based or workflow-based agent behavior.
Goal
Produce a pack that is:
- compact at the top level;
- layered by role or workflow;
- explicit about boundaries and escalation;
- easy to audit and extend.
Workflow
- Define the pack scope
- name the product, team, or operating surface;
- list who the pack is for;
- separate runtime-assistant use from development-agent use if both exist.
- Inventory shared vs specialized behavior
- identify what every role shares;
- identify what changes by role, workflow, or review depth;
- avoid creating separate skills when a difference layer would be enough.
- Choose the pack structure
- keep one directory per skill;
- keep
SKILL.mdshort; - move detailed rules, examples, and schemas into
references/.
- Model inheritance explicitly
- define the base workflow skill first;
- define supervisory or wider-scope skills as difference layers on top of the base;
- document when a user should switch to a narrower or wider mode.
- Define boundaries
- list what the assistant can do;
- list what it must escalate, refuse, or leave to humans;
- separate guidance, draft help, and verification from state-changing authority.
- Define rollout order
- start with the smallest set that proves the pattern;
- prefer 2-3 core skills before a full role matrix;
- add later skills only after the first pack is stable in use.
- Produce the design summary
- pack purpose;
- skill list;
- private vs reusable notes;
- known gaps and next additions.
Design Rules
- Base workflow first, difference layers second.
- One skill should answer one operating question clearly.
- Prefer references over long top-level instructions.
- Do not smuggle product policy into a supposedly generic methodology skill.
Final Report
Return the proposed pack structure, the first skills to build, the difference-layer model, and the rollout sequence.
What ships with it: 2 files
1.7 KB alongside SKILL.md
agents/
- openai.yaml228 B
- SKILL_CARD.md1.5 KB