dod
Project Definition of Done verification against the Deliverables Manifest — verify every deliverable, run tests, check VRTM, produce pass/fail report
Skill by Wave-Engineering
Chalk Labs
Need help deploying dod?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/wave-engineering-dod
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: documents
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/dod
What's inside
- Skill name
dod- Description
- Project Definition of Done verification against the Deliverables Manifest — verify every deliverable, run tests, check VRTM, produce pass/fail report
- License
- MIT
- Instructions
- 47 lines in SKILL.md · view on GitHub
Files (2)
- SKILL.md
- introduction.md
SKILL.md (first lines)
<!-- introduction-gate: If introduction.md exists in this skill's directory AND
the marker file /tmp/.skill-intro-dod does NOT exist, read introduction.md,
present its contents to the user, then create the marker: touch /tmp/.skill-intro-dod
Do NOT delete introduction.md — it lives in a protected directory.
Do this BEFORE executing any skill logic below. -->
# Project DoD Verification
Read the Deliverables Manifest from the project's Dev Spec and mechanically verify every deliverable exists at its declared path, every test passes, and VRTM is complete. Generate a pass/fail report and require explicit human sign-off before any campaign state change.
## Tools Used
- `mcp__sdlc-server__dod_load_manifest` — load and parse the Deliverables Manifest + Section 7 DoD + VRTM from a Dev Spec
- `mcp__sdlc-server__dod_verify_deliverable` — run the per-category verification (Docs / Code / Test / Trace)
- `mcp__sdlc-server__dod_run_test_suite` — execute the project's test target and return pass/fail summary
- `mcp__sdlc-server__dod_check_coverage` — parse the coverage report at the declared path
## Commands
`/dod` or `/dod check` — run the full verification.
## Procedure
1. **Locate the Dev Spec.** Use a user-provided path, else find `docs/*-devspec.md`. Multiple → ask. None → "No Dev Spec found. Run `/devspec create` first."
2. **Load the manifest.** Call `dod_load_manifest(prd_path)` to parse Sections 5.A (Deliverables), 7 (global DoD), 8 (phase DoD), and 9/Appendix V (VRTM). Separate active rows from `N/A -- <rationale>` rows (N/A without rationale is flagged).
3. **Verify each active deliverable.** For each row, call `dod_verify_deliverable(row)`. The tool applies the per-category rules:
- **Docs**: file exists and is non-empty
- **Code (binary/package)**: file exists; run discovered build if any
- **Code (CI/CD)**: workflow file exists and last CI run passed (via `gh run list` / `glab ci list`)
- **Code (build system)**: Makefile/task runner exists and `make test` (or equivalent) passes
- **Test (results)**: results file exists (e.g. `reports/junit.xml`) with passing counts — call `dod_run_test_suite()` if missing and regenerable
- **Test (coverage)**: `dod_check_coverage(path)` — file existence is the gate; percentage is informational
- **Test (manual procedures)**: document exists and has execution evidence
- **Trace (VRTM)**: every requirement traced, no `Pending` rows, all `Verified By` populated
4. **Check global DoD (Section 7).** Verify each item — phase DoD from Section 8 all checked, test plan items executed, cross-cutting conditions evidence in the codebase.
5. **Check VRTM completeness separately.** Every requirement ID from Section 3 appears; no `Pending` status; no empty `Verified By`.
6. **Present the verification report.** Header: project / Dev Spec path / date. Then `Deliverables Manifest: X/Y verified` with one row per deliverable using V (passing) / X (failing) / O (N/A). Then `Global DoD: X/Y` with item-level results. End with `RESULT: READY` or `RESULT: NOT READY -- N items failing`.
7. **Approval flow.** All pass → "Project DoD verified. Approve to close? (yes/no)". Failures → "N items failing. Approve anyway, or fix first? (yes/no/fix)". On **yes**, if `.sdlc/` exists run `campaign-status stage-review dod` and suggest closing the parent epic. On **fix**, list each failing item with a specific, actionable remediation (file path / command / Dev Spec section). On **no**, "Deferred. Re-run `/dod` when ready."
## Non-Negotiables
- **Mechanical verification only** — file exists or it does not, tests pass or they do not. No "looks good enough" judgments.
- **N/A is not a failure** — but bare `N/A` without a rationale is flagged.
- **VRTM completeness is mandatory** — `Pending` rows are failures. No exceptions.
- **Human approval is required** — even on all-green, present the report and wait.
- **Remediation is actionable** — specific file paths, commands, Dev Spec sections; not vague advice.
- **The Deliverables Manifest is the source of truth** — verify against Section 5.A, not assumptions.From Wave-Engineering/claudecode-workflow, MIT.
More from Wave-Engineering/claudecode-workflow
- 0.584score
assesswaves
Wave-Engineering
Quick assessment of whether a piece of work is suitable for wave-pattern execution (parallel or serial).
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
grunt
Wave-Engineering
Stand up a scoped-operations "grunt" — a dedicated aoe agent that clears a bounded service backlog so the main session's context stays clean for…
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
disc
Wave-Engineering
Discord integration — /disc routes to disc-server MCP tools, plus discord-watcher CLI subcommands for forward/directmsg/doorbell.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
multithread
Wave-Engineering
Parallelize a discussion over independent items — lay out all threads at once each with a take, batch-answer, drop sorted ones, iterate until dry.
SkillAFreshness: A★ 1Claude Code +9 - 0.581score
engage
Wave-Engineering
Read CLAUDE.md, confirm development rules of engagement, and load current plan
SkillAFreshness: A★ 1Claude Code +9 - 0.581score
ibm
Wave-Engineering
Verify Issue → Branch → PR/MR workflow compliance via MCP tool
SkillAFreshness: A★ 1Claude Code +9 - 0.581score
jfail
Wave-Engineering
Fetch and analyze a failed CI job (GitLab) or workflow run (GitHub)
SkillAFreshness: A★ 1Claude Code +9 - 0.581score
lazyriver
Wave-Engineering
Goal-seek loop — probe, judge sufficiency, steer, journal; emits a plan or answer
SkillAFreshness: A★ 1Claude Code +9 - 0.581score
autochain
Wave-Engineering
Plan a dependency-ordered chain of issues, then execute it autonomously on approval
SkillAFreshness: A★ 1Claude Code +9 - 0.581score
name
Wave-Engineering
Report or pick the agent's session identity (Dev-Name, Dev-Avatar, Dev-Team)
SkillAFreshness: A★ 1Claude Code +9 - 0.581score
edit
Wave-Engineering
Open a file or URL in a GUI editor (modification intent, async)
SkillAFreshness: A★ 1Claude Code +9 - 0.580score
ccfold
Wave-Engineering
Merge upstream CLAUDE.md template changes into the current project's CLAUDE.md, preserving project-specific content like Dev-Team and custom sections.
SkillAFreshness: A★ 1Claude Code +9
Signals
- Stars
- 1
- Forks
- 1
- Open issues
- 85
- Last commit (this folder)
- Jun 21, 2026 (4 months ago)
- Last release
- Oct 24, 2026 (6 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from Wave-Engineering
- 0.584score
grunt
Wave-Engineering
Stand up a scoped-operations "grunt" — a dedicated aoe agent that clears a bounded service backlog so the main session's context stays clean for…
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
multithread
Wave-Engineering
Parallelize a discussion over independent items — lay out all threads at once each with a take, batch-answer, drop sorted ones, iterate until dry.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
assesswaves
Wave-Engineering
Quick assessment of whether a piece of work is suitable for wave-pattern execution (parallel or serial).
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
disc
Wave-Engineering
Discord integration — /disc routes to disc-server MCP tools, plus discord-watcher CLI subcommands for forward/directmsg/doorbell.
SkillAFreshness: A★ 1Claude Code +9 - 0.581score
edit
Wave-Engineering
Open a file or URL in a GUI editor (modification intent, async)
SkillAFreshness: A★ 1Claude Code +9 - 0.581score
autochain
Wave-Engineering
Plan a dependency-ordered chain of issues, then execute it autonomously on approval
SkillAFreshness: A★ 1Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get dod running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
dod skill. Agent Mart. https://agentmart.dev/skills/wave-engineering-dod (last updated 2026-10-24)