principle-prove-it-works
Apply after completing a task, before declaring done. Verify against the real artifact (run the feature, read the actual value, inspect the diff), not a proxy, self-report, or "it compiles."
- License 许可
- MIT
- 所属仓库
- ravnhq/ravn-core
- 最近更新
- 2026-11-13
- Stars
- ★ 2
- Forks
- 0
01 前言 Front Matter
- name
- principle-prove-it-works
- metadata
- { "proof": "subjective" }
- description
- Apply after completing a task, before declaring done. Verify against the real artifact (run the feature, read the actual value, inspect the diff), not a proxy, self-report, or "it compiles."
- disable-model-invocation
- true
02 资源文件 3
- 参考文档agents/openai.yaml135 B
- 许可证LICENSE.txt1.1 KB
- 参考文档TEST-BEHAVIOR.md2.4 KB
03 SKILL.md 正文
Prove It Works
Verify every task output against the real thing, directly, before declaring it done. Ask "how do I prove this actually works?" Indirect checks (file mtimes, output freshness, self-reports, cached screenshots, "it compiles") feel cheaper, and acting on a wrong inference costs far more.
- Check process liveness directly, not through derived state.
- Read the actual value, not a cached or derived representation.
- When verification fails, suspect the observation method before the system.
Code and features:
- Build it (necessary but not sufficient)
- Run it and exercise the actual feature path
- Check the full chain: does data flow from input to output?
- For integrations, test the full communication path end-to-end
When writing, changing, or keeping a test, apply Test Behavior, Not Implementation.
Delegation: trust artifacts, not self-reports. Inspect the delegate's actual output (git diff, file contents, runtime behavior), not its summary.
Disclosure: report every change you made, including edits a check would reject and anything moved out of view by stash, reset, amend, or checkout. Never shape git state so a checker sees less than you did. A check that passes because the change is hidden proves nothing.
Script the check when you can
The strongest proof is a deterministic script that re-runs the same comparison. Run it and keep its output as an artifact a reviewer can re-run instead of trusting your word.
Keep the artifact visible for the human. Commit it only for large or complex work where the trail has to be auditable later, like a big port or migration.
04 同仓库更多 Skills
writing-for-agents
Writing documents for agents. Use when creating or editing skills, or modifying AGENTS.md or CLAUDE.md.
why
Use for 'why does X work this way', 'why we picked Y', design rationale, regressions, postmortems, or data-backed thresholds. Queries every available evidence MCP in parallel and returns a cited read on decisions and tradeoffs. Use how for runtime behavior.
typescript-best-practices
TypeScript type-safety rules: discriminated unions, branded types, no `as`, exhaustive switches, schema-derived types. Use when writing, editing, or reviewing TypeScript.
show-me-your-work
Keep a reviewable decision trail for unattended or multi-phase work: a TSV log with one row per decision (what, why, evidence, result).
ship
Take a change from idea to a ready PR with passing CI, running research, grilling, design, verification, and the pr skill in sequence.
session-pickup
Resume unfinished work from a handoff note or prior coding-agent session. Use before reading a supplied handoff, even when its next action is only an inspection or report. Also use to pick up, continue where you left off, or take over from a transcript, agent URL, or pushed branch. Use pause-safely to write a handoff.