| name | ed-discovery |
| description | Discover useful tools, concepts, or mental models that apply to real work problems. Like a well-read friend giving you a practical insight. Triggers on: discovery, discover, explore new, new tool, bizu, descoberta. |
| user-invocable | true |
Discovery — Practical Insight
Use this skill for open-ended exploration that should bring back something useful: a tool, concept, mental model, pattern from another industry, cultural idea, historical analogy, or emerging technique.
Unlike ed-research, the target does not need to be known in advance. The value is finding something the agent/operator would not naturally search for, then making its practical relevance clear.
Arguments
- No argument: explore freely and bring back something useful.
- With direction: explore in that direction while keeping room for adjacent discoveries.
Boundary
Do not manage lifecycle, publication, postflight, or generic artifact rites inside this skill.
What Counts As A Discovery
A good discovery has practical application, not just novelty.
Examples:
- Tool: "prompt tuning by hand can become an optimization loop with DSPy."
- Concept: "Andon cord maps to fail-fast pipeline interruption."
- Industry pattern: "aviation checklists clarify handoff risk in agent processes."
- Cultural concept: "genchi genbutsu means go see the real system, not the proxy."
Not enough: "this is interesting." The discovery must answer: what changes because we know this?
Method
1. Choose A Starting Direction
Start from one of:
- current work friction;
- a vague operator signal;
- a surprising thread in prior research;
- a nearby field;
- a tool ecosystem;
- an analogy from another discipline;
- pure curiosity when no stronger signal exists.
2. Explore Broadly
Search across sources appropriate to the direction:
- tool ecosystems and repositories;
- HN, technical blogs, docs, papers, and public discussions;
- other industries such as manufacturing, aviation, medicine, education, operations;
- history, philosophy, language, and cultural concepts.
The source itself can be the discovery if it reveals a reusable idea.
3. Understand The Discovery
For tools, capture:
- what it does;
- how it works;
- how to get started;
- cost and operating constraints;
- limitations and failure modes.
For concepts or patterns, capture:
- original context;
- core mechanism;
- why it mattered there;
- where the analogy does and does not transfer.
4. Contextualize To Work
This is the central requirement.
Explain concretely:
- which project or work stream it affects;
- what current friction it addresses;
- how things work now;
- how they would work with the discovery;
- what first practical step would test it;
- what risk would make the analogy fail.
Quality Criteria
- The application to work must be specific enough to act on.
- Include a before/after comparison for the main application.
- Explain the concept plainly before using it as advice.
- Distinguish useful analogy from overreach.
- Prefer one strong discovery over a list of loosely related curiosities.
- Keep sources traceable.
Output Contract
Produce a discovery artifact suitable for the uniform report pipeline.
Recommended sections:
- The Problem Or Friction
- The Discovery
- Original Context
- Application To Work
- Before And After
- Getting Started
- Risks And Limits
- References
Privacy Rule
For external posts or public communication, do not identify private organizations, owners, project names, or data that can trace the human operator.
