Files
magnus919_agent-skills/cncf-landscape/README.md
Magnus HedemarkandGitHub 622f341ea4 feat: add CNCF Landscape technology selection skill
Merged after successful CI and exact-head code review.
2026-08-02 02:06:38 -04:00

51 lines
2.7 KiB
Markdown
Raw Permalink Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
# CNCF Landscape
Discover cloud-native technologies and turn the result into a decision-ready shortlist with evidence, trade-offs, and a validation plan.
## Why Install This Skill
When an architecture question starts with “what exists for this?”, an agent can easily return a familiar-name list or rank projects by stars. This skill gives it a live, source-grounded discovery path through the CNCF Landscape and a disciplined way to separate catalog facts from engineering judgment.
It is useful for architects and engineers exploring a capability that is not yet in their stack. The bundled query tool handles the static APIs filtering and bounded JSON output; the skill then asks the questions the catalog cannot answer: who will operate it, what constraints matter, what evidence is missing, and what small experiment could falsify the recommendation.
## What You Get
| Path | Purpose |
|---|---|
| `SKILL.md` | Trigger boundaries, query workflow, evidence discipline, and completion criteria |
| `scripts/landscape_query.py` | Read-only stdlib CLI for live project/member queries and local filtering |
| `references/api.md` | Verified endpoint map, field semantics, and static-site caveats |
| `references/decision-framework.md` | Candidate comparison and recommendation method |
| `references/output-template.md` | Reusable decision artifact structure |
| `tests/test_landscape_query.py` | Offline client and filter tests |
| `evals/evals.json` | Six output-quality cases covering normal and failure paths |
| `evals/trigger-queries.json` | Three should-trigger and two should-not-trigger routing probes |
## Quick Start
Requires Python 3.8+ and outbound HTTPS access. No API key is required.
```bash
python3 scripts/landscape_query.py \
--category "Observability and Analysis" \
--subcategory Observability \
--search tracing \
--maturity graduated \
--has-license --has-release \
--sort stars --limit 10
```
The command emits a JSON envelope containing the source endpoint, retrieval time, filters, counts, and matching records. Ask an Agent Skills-compatible assistant to interpret that evidence against your workload and constraints rather than treating the result as an automatic ranking.
## Triggers
Use when discovering or comparing CNCF/cloud-native projects, building a shortlist for an architecture decision, filtering technology candidates by maturity or repository evidence, or investigating what tools exist for a capability missing from the current stack.
Do not use it for operating a named project, general architecture methodology, or legal/procurement conclusions.
## Requirements
- Python 3.8 or newer
- Network access to `https://landscape.cncf.io`
- No credentials or third-party Python packages