zeno-validating-work

Original：🇺🇸 English

Translated

Use when about to submit work, before committing or pushing, to run verification commands and confirm the work actually passes before claiming completion

2installs

Sourcezenon-red/skills

Added on2026-05-10

NPX Install

npx skill4agent add zenon-red/skills zeno-validating-work

SKILL.md Content

View Translation Comparison →

Validating Work (Contributor Workflow)

Overview

Run verification commands and confirm actual results before claiming work is complete or submitting. Evidence before assertions.

Trigger: About to commit, push, or claim task complete

Action: Run tests/verification → Confirm output → Then submit

The Gate Function

BEFORE claiming completion or submitting:

1. IDENTIFY: What proves this task is done?
   - Tests? Which command?
   - Linter? Which command?
   - Manual verification? What to check?

2. RUN: Execute the FULL command fresh
   - Don't trust previous runs
   - Run it now in this session

3. READ: Full output, check exit code, count failures

4. VERIFY: Does output confirm success?
   - YES: State completion WITH evidence
   - NO: Fix issues, re-run, or report actual status

5. ONLY THEN: Commit, push, or claim complete

Skip any step = unverified claim

Verification Commands by Component

stdb (Rust):

bash

cd stdb && cargo test && cargo clippy -- -D warnings
# Expected: All tests pass, no warnings

backend (Deno):

bash

cd backend && deno task test && deno task lint
# Expected: All tests pass, no lint errors

frontend (Node):

bash

cd frontend && npm run lint:all && npm run typecheck
# Expected: No lint errors, no type errors

docs / text contributions:

bash

# Content quality self-check before submitting:

| Check | Pass Means |
|-------|------------|
| Task complete | Every item in the task description is addressed |
| No filler | No "In this document we will..." fluff, no AI padding |
| Facts verified | Every claim about a repo, API, or feature checked against its source |
| Atomic scope | One PR = one task. No bundling unrelated work |
| Evidence | Command output, screenshots, or test results included |
| Readable | No obvious spelling errors, coherent structure |

Common Verification Needs

Task Type	Verify	Command Example
Feature	Tests pass	`cargo test feature_name`
Bug fix	Regression test + original symptom fixed	Run reproduction steps
Refactor	No behavioral changes	Full test suite
Docs	Links/formatting	Manual check or link checker

Anti-Patterns

❌ Wrong: "Should work now" without running tests ✅ Right:

[Run tests] [See: 12/12 pass] "Tests pass, ready to submit"

❌ Wrong: "Linter passed" for compile check ✅ Right: Separate verification:

cargo check

then

cargo clippy

❌ Wrong: Claiming completion while tests failing ✅ Right: Report actual status: "3 tests failing, need to fix X"

❌ Wrong: Trusting a previous test run ✅ Right: Fresh verification in current session

Red Flags - STOP

About to push without running verification
Using "should", "probably", "seems" before verification
Expressing satisfaction before checking: "Done!" (without verifying)
Thinking "just this once" skip verification

Integration

Before:

executing-tasks

(completing the work) After:

submitting-work

(push and create PR)

Bottom Line

Run verification. Read output. Then claim completion.

No shortcuts. No assumptions. Evidence first.

zeno-validating-work

NPX Install

Tags

SKILL.md Content

Validating Work (Contributor Workflow)

Overview

The Gate Function

Verification Commands by Component

Common Verification Needs

Anti-Patterns

Red Flags - STOP

Integration

Bottom Line