sdd
Skill0 starsChanged 29 days agoApache-2.0View on GitHub
Use when starting or resuming any feature with Spec-Driven Development, or when unsure which SDD phase comes next. Explains the cycle and routes to sdd-specify, sdd-plan, sdd-tasks, sdd-implement or sdd-verify based on which files exist.
20 other projects use a version of this.
Adopted by
Most recently taken up by
- AsiaOstrich/EngramGraphtheir own version, first seen Nov 8, 2026
- AsiaOstrich/universal-dev-standardstheir own version, first seen Nov 1, 2026
- totalfrank/skillstheir own version, first seen Oct 27, 2026
- Isabella-a/Skillstheir own version, first seen Oct 19, 2026
- IulaiJedi/sdd-skilltheir own version, first seen Oct 18, 2026
- artimath/effect-skillstheir own version, first seen Oct 17, 2026
Used with
Projects with it also use bdd, migrate and 3 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
TashanGKDSkillbdd
行为驱动开发(BDD)流程助手。用 Given-When-Then 描述验收标准,包含 Feature 文件管理、步骤定义实现。用户提到 BDD、验收测试、Gherkin 时使用。
Also in 3 projects that use this
tuistSkillmigrate
Migrates existing Xcode projects to Tuist generated workspaces with build and run validation, external dependency mapping, and migration checklists. Use when adopting Tuist for an existing app or converting a hand-edited Xcode project to generated projects.
Also in 4 projects that use this
BlockSkillplan
Draft a TDD, PRD, or lightweight plan in the conversation without committing it.
Also in 3 projects that use this
Hugging FaceSkillimplement
Make tests pass. Invoke after /write-tests produces failing tests.
Also in 3 projects that use this
GoogleSkilltdd
Test-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.
Also in 4 projects that use this
Other versions20 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 20
- siemens/ix★ 390same name
- SpillwaveSolutions/sdd-skill★ 99same name
- liatrio-labs/spec-driven-workflow★ 92same name
- AsiaOstrich/universal-dev-standards★ 84same name
- baekenough/oh-my-customcode★ 42same name
- dewil/ai-toolkit★ 9same name
- dewil/claude-toolkit★ 9same name
- artimath/effect-skills★ 8same name
- AsiaOstrich/EngramGraph★ 4same name
- kingdon/skills★ 4same name
10 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository7 other files it gives its agents.
- AGENTS.mdAGENTS.md
- sdd-constitutionSkill
- sdd-implementSkill
- sdd-planSkill
- sdd-specifySkill
- sdd-tasksSkill
- sdd-verifySkill
The file
We split this skill into the rules it gives an agent
7 rules. 0 are common, 1 is distinctive, and 6 are very specific to this project.
0
Common
Most of the 11 edited versions we read keep these.
1
Distinctive
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 11
Never keep the state of a feature only in the conversation.
- Very specific: line 29
Read .sdd/config.yaml (project, else ~/.sdd/config.yaml) for specs_dir, gates and phases.
- Very specific: line 33
If the user describes something new, start a new feature with sdd-specify.
- Very specific: line 50
After a FAIL: If verify.md says Verdict: FAIL, do its "Next actions" (they send you back to specify, tasks or implement).
- Very specific: line 58
Human gate: After each phase listed in gates (default specify, plan, tasks), show the artifact and stop until the user approves.
- Very specific: line 62
Traceability: Every plan step and every task cites at least one REQ-n.
- Very specific: line 66
Spec first: When work reveals a new or changed requirement, it goes into spec.md as a new REQ-n before any code changes.
“Common” is measured: kept by at least 6 in 10 of the 11 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Spec-Driven Development (SDD)
Work flows spec → plan → tasks → implement → verify. Every phase leaves a file on disk. Those files are the shared memory between sessions, people and agents: anyone — you in a fresh session, a teammate, a different coding agent — can pick the work up by reading them. Never keep the state of a feature only in the conversation.
Artifacts
specs/
constitution.md project principles (optional, once per repo)
NNN-<slug>/
spec.md REQ-1… what and why; verifiable acceptance criteria
plan.md decisions, files touched, risks; each step cites REQ-n
tasks.md T-1… checklist; each task cites REQ-n and names its check
verify.md REQ → task → evidence matrix; uncovered items flaggedspecs/ and the REQ/T prefixes are defaults; .sdd/config.yaml can change them.
Where am I? (routing)
- Read
.sdd/config.yaml(project, else~/.sdd/config.yaml) forspecs_dir,gatesandphases. Missing file = defaults:specs_dir: specs,gates: [specify, plan, tasks],phases: [constitution, specify, plan, tasks, implement, verify]. - Pick the feature dir: the one the user names, else the highest
NNN-*inspecs_dir. If the user describes something new, start a new feature withsdd-specify. - Walk
phasesin order. The next step is the first phase inphasesthat is not done:
| Phase | Done when | Skill |
constitution | always (optional — suggest it once if <specs_dir>/constitution.md is missing, never block on it) | sdd-constitution |
specify | spec.md exists | sdd-specify |
plan | plan.md exists | sdd-plan |
tasks | tasks.md exists | sdd-tasks |
implement | tasks.md has no - [ ] left | sdd-implement |
verify | verify.md says Verdict: PASS | sdd-verify |
| any other name | the file its skill says it writes exists | sdd-<name>, else <name> |
Approval lives in the file. A phase listed in gates is only done when its artifact's
**Status:** line says approved. If the artifact exists but its status is not approved,
the next step is that phase's gate: show the artifact and ask for approval — do not move on.
After a FAIL. If verify.md says Verdict: FAIL, do its "Next actions" (they send
you back to specify, tasks or implement). Once they are done and every task is ticked
again, run sdd-verify again; it rewrites verify.md.
Tell the user where the feature stands (one line), then use the next skill.
Rules that hold in every phase
Human gate. After each phase listed in gates (default specify, plan, tasks),
show the artifact and stop until the user approves. Approval of one phase is not approval
of the next.
Traceability. Every plan step and every task cites at least one REQ-n. A requirement
nobody cites is not going to be built; a task citing nothing is scope creep. sdd-verify
fails both.
Spec first. When work reveals a new or changed requirement, it goes into spec.md as a
new REQ-n before any code changes. IDs are permanent: never renumber, never reuse.
One task at a time. Implement a task, run its check, tick it, commit. Then the next.
Customizing
.sdd/config.yaml— paths, language, ID prefixes, gates, verify command, phase order..sdd/templates/<name>.md— replace any shipped template..sdd/rules/all.md,.sdd/rules/<phase>.md— add your own rules to every phase or one.- Guide: https://github.com/monaccode/astromesh-sdd/blob/main/docs/extending.md
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

