sdd-init
Detect project stack, testing capabilities, and initialize SDD context. Trigger: When initializing SDD, iniciar sdd, openspec init, /sdd-init
- License 许可
- MIT
- 最近更新
- 2026-10-22
- Stars
- ★ 1
- Forks
- 0
01 前言 Front Matter
- name
- sdd-init
- scope
- core
- license
- MIT
- metadata
- { "author": "gentleman-programming", "version": "3.0" }
- description
- Detect project stack, testing capabilities, and initialize SDD context. Trigger: When initializing SDD, iniciar sdd, openspec init, /sdd-init
- display_name
- SDD Init
- user-invocable
- false
- disable-model-invocation
- true
02 资源文件 1
- 参考文档references/init-details.md6.4 KB
03 SKILL.md 正文
<!-- Synced from https://raw.githubusercontent.com/Gentleman-Programming/gentle-ai/v1.26.5/internal/assets/skills/sdd-init/SKILL.md (tag v1.26.5, commit 5f73974b39ae2b9b525ef465b3642030c5f2ce6c); adapted for Jarvis/Hive runtime semantics. -->
Activation Contract
Run this phase when the orchestrator/user asks to initialize SDD in a project. You are the phase executor: do the work yourself, do not delegate, and do not behave like the orchestrator.
Hard Rules
- Detect the real stack, conventions, architecture, testing tools, and persistence mode; never guess.
- In
hivemode, do not createopenspec/. - In
openspecmode, follow../_shared/openspec-convention.mdand write file artifacts. - In
hybridmode, write both openspec files and Hive observations. - Always persist testing capabilities separately as
sdd/{project}/testing-capabilitiesoropenspec/config.yamltesting:. - A test runner counts as detected only when a real test command exists; a config or manifest file alone is not enough. Use the Testing Capability Checklist in
references/init-details.md. - Detection produces a suggestion, not a decision: cache
strict_tdd_suggestion: strict|standardplus a one-linedetection_reason; the preflightTDD modedecides per feature. - Keep caching
strict_tddfor backward compatibility; it mirrors the suggestion only (truewhen the suggestion isstrict) and never activates Strict TDD by itself. - Always build
.jarvis/skill-registry.md; also saveskill-registryto Hive when available. - Use
capture_prompt: falsefor automated SDD/config saves when supported; omit it if the tool schema lacks it. - If
openspec/already exists, report what exists and ask before updating it. - Artifact store modes supported by Jarvis skills:
hive | openspec | hybrid | none.
Decision Gates
| Input | Action |
|---|---|
mode=hive |
Save context and capabilities to Hive only. |
mode=openspec |
Create/update openspec bootstrap files only. |
mode=hybrid |
Do both Hive and openspec persistence. |
mode=none |
Return detected context only; write no SDD artifacts except registry if required. |
| strict TDD marker/config found | Use that value as strict_tdd_suggestion. |
| no marker/config and project code is mainly Deluge | Suggest standard: Deluge code cannot run under a local test runner. |
| no marker/config and a real test command exists | Suggest strict. |
| no real test command | Suggest standard and record why in detection_reason. |
Execution Steps
- Inspect project files (
package.json,go.mod,pyproject.toml, CI, lint/test config) and summarize stack/conventions. - Detect real test commands, test layers, coverage, linter, type checker, and formatter.
- Resolve the Strict TDD suggestion in gate order: agent marker or
openspec/config.yamlstrict_tdd:, then Deluge, then real test command, then no test command. Record the one-linedetection_reason. - Initialize persistence for the resolved mode.
- Build
.jarvis/skill-registry.mdusing the skill-registry scan rules. - Persist testing capabilities and project context.
- Return the structured initialization envelope.
Output Contract
Return status, executive_summary, artifacts, next_recommended, and risks. Include project, stack, persistence mode, Strict TDD suggestion with its detection reason, testing capability table, saved observation IDs/paths, registry path, and next /sdd-explore or /sdd-new step.
References
- references/init-details.md — detection checklist, Hive payloads, config skeleton, and output templates.
../_shared/hive-convention.md— Hive artifact naming.../_shared/openspec-convention.md— openspec layout and rules.
04 同仓库更多 Skills
work-unit-commits
Plan commits as reviewable work units. Trigger: implementation, commit splitting, chained PRs, or keeping tests and docs with code.
skill-registry
Create or update the skill registry for the current project. Scans user skills and project conventions, writes .jarvis/skill-registry.md, and saves to Hive if available. Trigger: When user says "update skills", "skill registry", "actualizar skills", "update registry", or after installing/removing skills.
skill-improver
Trigger: improve skills, audit skills, refactor skills, skill quality. Audit and upgrade existing LLM-first skills safely.
skill-creator
Trigger: new skills, agent instructions, documenting AI usage patterns. Create LLM-first skills with valid frontmatter.
sdd-verify
Verify implementation against specs with structural and behavioral checks. Trigger: When verifying implementation
sdd-tasks
Break down a change into a concrete, ordered implementation checklist. Trigger: When creating task lists