MVPavan/via / .claude/skills/authoring-for-agents
authoring-for-agents
VerifiedUse when creating or changing agent instructions or their activation behavior. Skip edits that only correct spelling or formatting.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 2.4 KB |
| agents/openai.yaml | config | 42 B |
| references/skill-anatomy.md | markdown | 1.8 KB |
| references/testing-docs.md | markdown | 1.4 KB |
| references/writing-principles.md | markdown | 1.8 KB |
references/writing-principles.md
View on GitHub# Writing principles
## Routing and load
A pointer should name the material and the condition for reading it. Front-load
the trigger and distinguish the nearest competing workflow. Descriptions can
consume discovery context even when the body stays unloaded; exact exposure
depends on the harness and invocation metadata.
Keep common steps inline. Move substantial branch-specific material behind a
conditional pointer. Avoid chains of mandatory reads and duplicate policy owners.
Manual entrypoints trade automatic discovery cost for a name the user remembers;
retain useful aliases as thin pointers rather than independent workflows.
## Precision and discretion
Use observable conditions, explicit outputs and checkable completion criteria.
State authority boundaries where a workflow could otherwise expand the task.
Match prescription to fragility: exact commands for deterministic contracts,
principles for choices that depend on context. Permit genuine exceptions with
clear conditions rather than appending vague qualifications to every rule.
Positive target behavior often helps explain a desired output; explicit
prohibitions remain appropriate for consequential boundaries. Neither form is
universally superior. Validate wording against the behavior at issue.
## Pruning
Keep what changes decisions: repository conventions, non-obvious failure modes,
required interfaces and operational invariants. Link to authoritative commands
and configuration instead of copying volatile facts. Remove redundant steps,
unsupported psychological claims and invented numerical thresholds. Familiar
terms can compress an explanation; coined labels need definitions and maintenance.
Judge savings by the full path an agent reads and the work it performs, not just
entrypoint length. A shorter prompt that loses a necessary check creates rework.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |