Agent skill

Nw Execute

by nWave-ai in nWave-ai/nWave

A skill your agent uses when a DELIVER roadmap already exists and you need to dispatch exactly one identified step through its TDD cycle.

MITAuto-check passedTesting & QA

Install Nw Execute

skills CLI
$ npx skills add nWave-ai/nWave --skill nw-execute -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install nWave-ai/nWave nw-execute --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/nWave-ai/nWave.git skills-src && mkdir -p .claude/skills && cp -r skills-src/nWave/skills/nw-execute .claude/skills/nw-execute && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
nw-execute
GitHub stars
617
Token cost
~3k tokens
SKILL.md length
528 words
Files
1
Skills in repo
14
Repo updated
First seen
Licence
MIT

At a glance

A skill your agent uses when a DELIVER roadmap already exists and you need to dispatch exactly one identified step through its TDD cycle.

  • Works in 5 steps: Parse Parameters — Extract agent name,… → Load Rigor Profile — Read… → Validate Context Files — Confirm… → …
  • A DELIVER roadmap already exists and you need to dispatch exactly one identified step through its TDD cycle
  • SKILL.md covers Overview, Syntax, Context Files Required and Rigor Profile Integration, plus 8 more sections
  • Calls python

What it does

Nw Execute is an agent skill from nWave-ai/nWave. Use when a DELIVER roadmap already exists and you need to dispatch exactly one identified step through its TDD cycle. Use nw-roadmap to create the plan, nw-deliver for the whole wave, and nw-continue to resume at the next inferred step.

Its SKILL.md is about 3k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Testing & QA, covering Test-driven development. The repository describes itself as: AI agents that guide you from idea to working code, with you in control at every step. The licence is MIT.

When your agent uses it

  • A DELIVER roadmap already exists and you need to dispatch exactly one identified step through its TDD cycle
  • Tasks that involve Test-driven development

Example prompts

  • “/nw-execute”

Requirements

  • Python 3

Workflow steps

5 steps, taken from the first numbered list in SKILL.md.

  1. Parse Parameters — Extract agent name, feature ID, and step ID from invocation. Gate: all three parameters present and non-empty.
  2. Load Rigor Profile — Read .nwave/des-config.json key rigor (default: standard if absent). Gate: config loaded or default applied.
  3. Validate Context Files — Confirm roadmap.json and execution-log.json exist under docs/feature/{feature-id}/deliver/. Gate: both files…
  4. Extract Step Context — Grep roadmap for step_id: "{step-id}" with ~50 lines context. Gate: step found; report available step IDs if missing.
  5. Invoke Agent — Call Agent tool with DES template below, applying rigor model and phases from step 2. Gate: Agent tool called, not executed…

What it can do on your machine

Read from SKILL.md and the folder at commit da401a8. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • python

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Nw Execute loads about 3k tokens when it runs. Until then it costs about 62 tokens; SKILL.md has 528 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~62
When it runs · the whole SKILL.md, loaded when a task matches
~3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from nWave-ai/nWave at commit da401a8, republished under its MIT licence (© nWave-ai). 528 words, ~3,018 tokens.

Download SKILL.mdSave it as .claude/skills/nw-execute/SKILL.md (or your agent's skills folder).
name
nw-execute
description
Use when a DELIVER roadmap already exists and you need to dispatch exactly one identified step through its TDD cycle. Use nw-roadmap to create the plan, nw-deliver for the whole wave, and nw-continue to resume at the next inferred step.
user-invocable
true
argument-hint
[agent] [feature-id] [step-id] - Example: @nw-software-crafter "auth-upgrade" "01-01"

NW-EXECUTE: Atomic Task Execution

Wave: EXECUTION_WAVE | Agent: Dispatched agent (specified by caller)

Overview

Dispatch one unit of DELIVER work to an agent: a single roadmap step. /nw-execute extracts the step from roadmap.json and dispatches it for the 3-phase TDD canon; the agent appends phase events to execution-log.json.

Syntax

/nw-execute @{agent} "{feature-id}" "{step-id}"

Context Files Required

  • docs/feature/{feature-id}/deliver/roadmap.json — Orchestrator reads once, extracts step context
  • docs/feature/{feature-id}/deliver/execution-log.json — Agent appends only (never reads)

Rigor Profile Integration

Before dispatching the agent, read rigor config from .nwave/des-config.json (key: rigor). If absent, use standard defaults.

  • agent_model: Pass as model parameter to Agent tool. If "inherit", omit model (inherits from session).
  • tdd_phases: Modify the TDD_PHASES section in the DES template to match the configured phases. The 3-phase canon (ADR-025) is [RED, GREEN, COMMIT]; the lean variant is [RED, GREEN]. Legacy 5-phase contract ([PREPARE, RED_ACCEPTANCE, RED_UNIT, GREEN, COMMIT]) and its lean variant ([RED_UNIT, GREEN]) remain supported for audit-log replay of pre-2026-05-07 commits. Remove omitted phases' instructions from the template.
  • refactor_pass: If false, skip COMMIT phase refactoring instructions.

Dispatcher Workflow

  1. Parse Parameters — Extract agent name, feature ID, and step ID from invocation. Gate: all three parameters present and non-empty.
  2. Load Rigor Profile — Read .nwave/des-config.json key rigor (default: standard if absent). Gate: config loaded or default applied.
  3. Validate Context Files — Confirm roadmap.json and execution-log.json exist under docs/feature/{feature-id}/deliver/. Gate: both files present; report path-not-found if missing.
  4. Extract Step Context — Grep roadmap for step_id: "{step-id}" with ~50 lines context. Gate: step found; report available step IDs if missing.
  5. Invoke Agent — Call Agent tool with DES template below, applying rigor model and phases from step 2. Gate: Agent tool called, not executed inline.

Agent Invocation

@{agent}

Use this DES template verbatim. Fill {placeholders} from roadmap. Without DES markers, hooks cannot validate.

<!-- DES-VALIDATION : required -->
<!-- DES-PROJECT-ID : {feature-id} -->
<!-- DES-STEP-ID : {step-id} -->

# DES_METADATA
Step: {step-id}
Feature: {feature-id}
Command: /nw-execute

# AGENT_IDENTITY
Agent: {agent-name}

# SKILL_LOADING
Before starting TDD phases, read your skill files for methodology guidance.
Skills path: ~/.claude/skills/nw-{skill-name}/SKILL.md
Always load before RED: tdd-methodology.md, quality-framework.md (3-phase canon, ADR-025) — legacy 5-phase logs reference loading at PREPARE.
Load on-demand per phase as specified in your Skill Loading Strategy table.

# TASK_CONTEXT
{step context from roadmap - name|criteria|test_file|scenario_name|implementation_notes|deps|files_to_modify (per nWave/templates/roadmap-schema.json)}

# DESIGN_CONTEXT
{Summary of architectural decisions relevant to this step, extracted by the orchestrator from docs/product/architecture/brief.md and wave-decisions.md. Include: component structure, dependency boundaries, technology choices, and any design constraints that affect implementation. If no design artifacts exist, write "No design artifacts available — use project conventions."}

# TDD_PHASES
3-phase canon (ADR-025, 2026-05-07). Execute in order:

1. RED - Activate the pre-authored acceptance test (PRIMARY TBU DEFENSE); write PBT unit tests ONLY if the AT cannot reach GREEN without them.
   AT activation: If TASK_CONTEXT includes test_file, locate it and remove the @skip/@ignore/@pending/xit/.skip/[Ignore] marker from the target scenario (the AT scaffold was authored by DISTILL — DELIVER does NOT re-author ATs). Run it — must fail for business logic reason (not import/syntax error). Fail-for-right-reason gate: collected ≥ 1, failures ≥ 1, no collection errors, semantic AssertionError / expected-exception-not-thrown.
   PORT-TO-PORT PRINCIPLE: The acceptance test exercises the scenario through
   the driving port (application service, orchestrator, CLI handler, API controller),
   not a decomposed helper or internal class. A correctly-written port-to-port test
   makes TBU structurally impossible — if a new function were missing or unwired,
   THIS test stays RED. That is the entire point: GREEN is unreachable without wiring.
   Litmus test: "If I delete the call-site that wires the new code, does this test fail?"
   If no → the test is at the wrong level. Stop and flag to orchestrator (DISTILL re-author needed).
   Conditional unit-test authoring: write PBT unit tests (or integration tests for adapter/infrastructure code — adapters use real infrastructure, never mocked unit tests) ONLY when the AT requires them to reach GREEN. If the AT can pass via direct minimal implementation, skip unit-test authoring inside RED.

2. GREEN - Minimal code to pass AT + any unit tests authored in RED.
   After GREEN: run FULL test suite. If all pass, proceed to COMMIT immediately.
   Smell test: if any new function is only called from test code, your acceptance
   test is at the wrong abstraction level — stop and flag.
   Never move to new task or stop without committing green code.

3. COMMIT - Commit this step's owned files via `des-commit` (parallel-safe).
   Use `des-commit` instead of raw `git add` / `git commit`. It holds an
   exclusive lock and commits ONLY the paths you pass, so a parallel agent's
   staged work is never swept into your commit (issue #51 / ADR-027). Pass
   EVERY file you created or modified for this step (production + tests) as
   `--owned-paths`; anything omitted is not committed.

des-commit
--owned-paths <all files this step created/modified>
--step-id {step-id}
--task-id {project_id}
--message "feat(feature-id): implement feature X"

Pass the bare `project_id` value (for example `44`, not `#44` and not the
step id). The `Step-Id: {step-id}` and `Task-Id: {project_id}` trailers
required for DES verification are appended automatically in one git trailer
block. Resulting commit for project 44:

feat(feature-id): implement feature X

Step-Id: 02-01 Task-Id: 44


LEGACY 5-PHASE CONTRACT (ADR-024 era, pre-2026-05-07): PREPARE → RED_ACCEPTANCE → RED_UNIT → GREEN → COMMIT. Preserved for audit-log replay only — new work uses the 3-phase canon above. Audit-log entries referencing RED_ACCEPTANCE/RED_UNIT/PREPARE represent merged sub-steps now folded into RED.

# QUALITY_GATES
- All tests pass before COMMIT
- No skipped phases without blocked_by reason
- Coverage maintained or improved

# OUTCOME_RECORDING
Immediately after ACTUALLY EXECUTING each phase — and BEFORE any
permission-gated or interruption-prone command (e.g. `nix-build`, a network
fetch, a long verification run) — record it via the DES CLI. Do NOT batch
`des-log-phase` calls to end-of-run: a permission prompt or STOP between phases
drops the un-written entries, leaving an incomplete trace (e.g. 3/5) that fails
`des-verify-integrity` at finalize and forces a resume to re-log work already
done.

 des-log-phase \
   --project-dir docs/feature/{feature-id}/deliver \
   --step-id {step-id} \
   --phase {PHASE_NAME} \
   --status EXECUTED \
   --data PASS

For `--status EXECUTED`, `--data` MUST be exactly `PASS` or `FAIL` — no prose,
no prefix. The stop-hook StepCompletionValidator compares the value against
`{PASS, FAIL}` exactly; anything else (e.g. `--data "PASS: unit layer folded
into shape suite"`) is rejected as an INCOMPLETE_PHASE and the phase must be
re-logged. Put any explanation in your report to the orchestrator, not in
`--data`. Only `--status SKIPPED` may carry text — and only as a valid prefix
(see below).

For SKIPPED phases (genuinely not applicable):

 des-log-phase \
   --project-dir docs/feature/{feature-id}/deliver \
   --step-id {step-id} \
   --phase {PHASE_NAME} \
   --status SKIPPED \
   --data "NOT_APPLICABLE: reason"

CLI enforces real UTC timestamps and validates phase names.
Do NOT manually edit execution-log.json.
Use the DES CLI to record phase outcomes and create log files.
Python resolution: `$(command -v python3 || command -v python)` — works on macOS (python3 only), Linux, and Windows.

CRITICAL: Only the executing agent calls the CLI.
Orchestrator MUST NEVER write phase entries — only the agent that performed the work. A log entry without actual execution is a **violation that DES detects and that will cause integrity verification to fail**, blocking finalize.

# RECORDING_INTEGRITY
Valid Skip Prefixes: NOT_APPLICABLE, BLOCKED_BY_DEPENDENCY, APPROVED_SKIP, CHECKPOINT_PENDING
Anti-Fraud Rules:
- NEVER write EXECUTED for phases you did not actually perform
- NEVER invent timestamps — DES CLI generates real UTC timestamps
- DES audits all entries; integrity violations block finalize

# BOUNDARY_RULES
- Only modify files listed in step's files_to_modify
- Do not load roadmap.json
- Do not modify execution-log.json structure (append only)
- NEVER write execution-log entries for phases you did not execute

# TIMEOUT_INSTRUCTION
Target: 30 turns max. If approaching limit, COMMIT current progress.
If GREEN complete (all tests pass), MUST commit before returning — even at turn limit.

Configuration:

  • subagent_type: extracted agent name
  • Turn limits are defined in each agent's maxTurns frontmatter field (not as a tool parameter)
Show full SKILL.md (209 more words)Show less

Error Handling

  1. Invalid Agent — Report available agents from the agent registry. Gate: error message returned, no invocation attempted.
  2. Missing Context Files — Report exact path not found for roadmap or execution-log. Gate: clear path reported.
  3. Step Not in Roadmap — Report available step IDs from roadmap. Gate: list of valid IDs returned.
  4. Dependency Failure — Explain which blocking tasks are incomplete. Gate: blocking step IDs named explicitly.

Resume vs Restart

When subagent times out:

Last Completed Phase (3-phase canon)Legacy phase (5-phase)ActionRationale
GREEN (or later)GREENResumeOnly COMMIT remains (~5 turns)
RED with partial GREENRED_UNIT with partial GREENResumePreserves implementation progress
RED only (pre-GREEN)PREPARE or RED_ACCEPTANCERestartLittle context worth replaying

Resume costs ~50% more tokens/call due to context replay (measured: 3.7K vs 2.5K tokens/call). For <5 remaining turns, resume is efficient. For 15+ turns, restart is cheaper.

Examples

bash
/nw-execute @nw-software-crafter "des-us007-boundary-rules" "02-01"
/nw-execute @nw-researcher "auth-upgrade" "01-01"
/nw-execute @nw-software-crafter "des-us007" "03-01"  # retry after failure

TDD_PHASES

<!-- Schema v4.0 — canonical source: TDDPhaseValidator.MANDATORY_PHASES -->
<!-- Build system injects mandatory phases from step-tdd-cycle-schema.json -->

{{MANDATORY_PHASES}}

Success Criteria

  • Agent invoked via Agent tool (dispatcher does not execute the work)
  • Step context extracted from roadmap and passed in prompt
  • Agent appended phase events to execution-log.json
  • Agent did not load roadmap.json

Next Wave

Handoff To: /nw-review for post-execution review Deliverables: Updated execution-log.json, implementation artifacts, and git commits

© nWave-ai, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in nWave/skills/nw-execute of nWave-ai/nWave.

Open the folder on GitHubat commit da401a8

Compare with similar skills

Nw Execute next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Nw Execute compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Nw Execute this skillnWave-ai/nWave617—~3kAutomated safety check: PassMIT
TDDpietheinstrengholt/rssmonster56430 repos~906Automated safety check: PassMIT
TDD WorkflowhellangleZ/burn-in-cceverywhere-ralph11211 repos~2.4kAutomated safety check: PassNone
TDDsanity-io/sanity6.4k20 repos~1kAutomated safety check: PassMIT
Test Driven Developmentfarm-fe/farm5.6k51 repos~2.5kAutomated safety check: PassMIT
Tapd Story PipelineTencentBlueKing/bk-bcs841—~2.6kAutomated safety check: PassCustom licence

Similar skills

  • TDD

    pietheinstrengholt/rssmonster

    Test-driven development. An agent skill from pietheinstrengholt/rssmonster.

    564 GitHub starsUsed in 30 repos~906 tokens
    Testing & QAAuto-check passed
  • TDD Workflow

    hellangleZ/burn-in-cceverywhere-ralph

    A skill your agent uses when writing new features, fixing bugs, or refactoring code.

    112 GitHub starsUsed in 11 repos~2.4k tokens
    Testing & QAAuto-check passed
  • TDD

    sanity-io/sanity

    Official

    Test-driven development with red-green-refactor loop. An agent skill from sanity-io/sanity.

    6.4k GitHub starsUsed in 20 repos~1k tokens
    Testing & QAAuto-check passed
  • A skill your agent uses when implementing any feature or bugfix, before writing implementation code

    5.6k GitHub starsUsed in 51 repos~2.5k tokens
    Testing & QAAuto-check passed
  • Tapd Story Pipeline

    TencentBlueKing/bk-bcs

    单需求实现流水线——把一个 TAPD 需求从零推进到代码提交。自动串联技术澄清、 开发计划、任务拆分、TDD 实现、架构/安全校验、代码提交六个阶段。

    841 GitHub stars~2.6k tokensUpdated yesterday
    Testing & QAAuto-check passed
  • Absolute Init

    maddhruv/absolute

    One-time setup for absolute: interview how you want it to behave (output style, autonomy, TDD strictness, spec dir, families) + detect the stack once, then write .absolute.config.json (project…

    218 GitHub starsUsed in 1 repo~3k tokens
    Testing & QAAuto-check passed

More from nWave-ai/nWave

All 14 skills in this repo
  • DELIVER wave orchestration workflow -- 9 phases from baseline to finalization.

    617 GitHub stars~974 tokensUpdated 23 days ago
    Auto-check passed
  • Nw Diagram

    nWave-ai/nWave

    Generates C4 architecture diagrams (context, container, component) in Mermaid or PlantUML.

    617 GitHub stars~640 tokensUpdated 23 days ago
    Auto-check passed
  • Nw Diverge

    nWave-ai/nWave

    Generates 3-5 divergent design directions through JTBD analysis, competitive research, structured brainstorming, and taste evaluation before convergence.

    617 GitHub stars~2.2k tokensUpdated 23 days ago
    Auto-check passed
  • Nw Document

    nWave-ai/nWave

    Creates evidence-based documentation following DIVIO/Diataxis principles.

    617 GitHub stars~1.4k tokensUpdated 23 days ago
    Auto-check passed
  • Nw Forge

    nWave-ai/nWave

    Creates new specialized agents using the 5-phase workflow (ANALYZE DESIGN CREATE VALIDATE REFINE).

    617 GitHub stars~626 tokensUpdated 23 days ago
    Auto-check passed
  • Nw Refactor

    nWave-ai/nWave

    Applies the Refactoring Priority Premise (RPP) levels L1-L6 for systematic code refactoring.

    617 GitHub stars~1.2k tokensUpdated 23 days ago
    Auto-check passed

Categories

Questions about Nw Execute

What does Nw Execute do?

A skill your agent uses when a DELIVER roadmap already exists and you need to dispatch exactly one identified step through its TDD cycle. Nw Execute is an agent skill from nWave-ai/nWave. Use when a DELIVER roadmap already exists and you need to dispatch exactly one identified step through its TDD cycle.

When should I use Nw Execute?

Nw Execute fits situations like: A DELIVER roadmap already exists and you need to dispatch exactly one identified step through its TDD cycle; tasks that involve Test-driven development.

How do I install Nw Execute in Claude Code?

Run `npx skills add nWave-ai/nWave --skill nw-execute -a claude-code`. Or copy the skill folder (nWave/skills/nw-execute in nWave-ai/nWave) into .claude/skills/nw-execute in your project. Claude Code loads it when a task matches its description.

How do I install Nw Execute in Codex?

Run `npx skills add nWave-ai/nWave --skill nw-execute -a codex`. Or copy the skill folder (nWave/skills/nw-execute in nWave-ai/nWave) into .agents/skills/nw-execute in your project. Codex loads it when a task matches its description.

Can I use Nw Execute in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add nWave-ai/nWave --skill nw-execute -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/nw-execute, .gemini/skills/nw-execute, .github/skills/nw-execute and .opencode/skills/nw-execute in your project.

What does Nw Execute need to run?

Going by SKILL.md and its folder, Nw Execute needs the command-line tools its instructions call (python). Our summary lists: Python 3.

Does Nw Execute access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Nw Execute safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Nw Execute use?

Nw Execute is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Nw Execute use?

About 3k tokens (SKILL.md is roughly 12k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Nw Execute?

Skills that share tags, products or a category with Nw Execute: TDD (pietheinstrengholt/rssmonster, 564 stars), TDD Workflow (hellangleZ/burn-in-cceverywhere-ralph, 112 stars), TDD (sanity-io/sanity, 6.4k stars) and Test Driven Development (farm-fe/farm, 5.6k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Nw Execute?

nWave-ai (a GitHub organization) maintains it in nWave-ai/nWave, which has 617 GitHub stars. The repository holds 14 skills in this directory. The repository was last updated on September 16, 2026.

Source: nWave-ai/nWave on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.