Plugin Claude Code
kamioj's personal SDD workflow plugin marketplace.
Plugin Claude Code
kamioj's personal SDD workflow plugin marketplace.
Plugin Claude Code
Spec-driven development plugin: research → ask → propose → HARD GATE → apply → verify → archive, plus a goal-driven autonomous loop (/spec:loop) — enforced by 5 hooks (TBD scan / approval gate / archive audit / verify reminder / loop driver), with scope-dispatched dev agents and a fresh-context verifier agent.
Instructions file
Claude Code instructions for kamioj/spec-workflow, covering claude.md, what this repo is, dev loop (no build/test — regenerate, load, run for real), 4. update local installs and manifests (four, roles differ).
Agent
Use PROACTIVELY when /spec:apply needs to implement code. Builds frontend (Vue / React / uni-app / Flutter / HTML) or backend (Java/Spring / Python / PHP / Node) per the approved proposal.md ## What, according to the scope the main loop specifies at dispatch. Cross-stack changes: the main loop dispatches TWO spec-dev…
Agent
Use PROACTIVELY when /spec:verify runs. Independent verification agent with a deliberately fresh context — never the conversation that implemented the change. Audits the diff against proposal.md's verify: clauses, the change's index.md (R-N requirement quotes + A-N assets), design.md contracts, and the coding charter…
Hook
Runs when you submit a prompt, before the agent sees it, executing check-tbd.sh, check-gate.sh and check-archive.sh via sh (3 commands). From kamioj/spec-workflow.
Hook
Runs when the agent finishes a response, executing loop-driver.sh and check-verify-reminder.sh via sh (2 commands). From kamioj/spec-workflow.
Skill Claude CodeCodex
Implement the code, advancing by proposal/tasks. A pre-command hook checks that proposal.md carries the required prerequisites. Verification model: cheap self-run working checks per node while implementing; ONE independent spec-verifier pass at the end (never dispatch verification subagents mid-implementation).
Skill Claude CodeCodex
Archives the current change to spec/archive/YYYY-MM-DD- /. Only invoked when the user explicitly says "archive". Writes an archive-stage retrospect, checks for uncommitted code, and is guarded by the check-archive hook.
Skill Claude CodeCodex
Interrogates preference-driven decision points. Works through each [TBD] in research.md one by one as numbered plain-text questions; answered items are moved to.
Skill Claude CodeCodex
Discussion mode. Everything said during this mode is treated as thinking material only — no documents are touched. Use for brainstorming directions, or to talk through concerns about a proposal before deciding how to change it.
Skill Claude CodeCodex
Spec-driven development workflow overview. Load this skill ONLY when the user explicitly asks for the spec flow — says "spec this first / draft a proposal / design first / write a proposal", or invokes a $spec-x command. NEVER self-activate on task size or complexity: size is at most grounds for a one-line suggestion…
Skill Claude CodeCodex
Technical design elaboration. Trigger on demand — use when the architecture is complex, there are more than 3 interfaces, sequence/data-flow diagrams are needed, or a decision requires more than 300 words of deep argumentation. Produces design.md.
Skill Claude CodeCodex
Streaming light tier for bug fixes and small changes (user-explicit only, never model-initiated). Appends F-N entries to the standing spec/changes/fixes/ batch — locate & confirm before touching code, fix directly when confident or research candidates when not, self-check evidence per entry; the ONE independent audit…
Skill Claude CodeCodex
Goal-driven autonomous round loop. Give it a goal + acceptance checklist + round budget (touchpoint 1), then it researches, implements, self-checks, and retrospects round after round — driven by a Stop-event hook, not by prompt discipline — until the final acceptance audit (touchpoint 2, the loop's ONLY independent…
Skill Claude CodeCodex
Write or fully rewrite proposal.md (## Why /.
Skill Claude CodeCodex
Researches the change — always maps the codebase status quo; spawns a web-survey researcher agent ONLY when the decision space is external (new dependency / tech choice / unfamiliar domain), never for internal refactors or fix batches. Findings go into research.md (single file; Open[TBD]/Decided maintained as before).…
Skill Claude CodeCodex
Resumes a paused change — deletes its .paused marker (after checking the single-active-change slot is free) and reports where the work left off so re-entry is instant.
Skill Claude CodeCodex
Edit a single proposal.md section (why / what / how / risk). A full rewrite goes through $spec-propose. After editing, the HARD GATE must run again.
Skill Claude CodeCodex
Installs the Codex-side SDD agent TOML files from the plugin bundle, reminds the user to trust hooks, and verifies that the SDD gates bite after plugin install.
Skill Claude CodeCodex
Closes the fix batch — dispatches ONE spec-verifier over the batch's accumulated diff (F-N entries are the claim anchors), writes the audit into fix.md, and archives spec/changes/fixes/ on pass; on fail the batch stays in place with findings recorded. Gated by check-archive (empty batch blocks; archiving a fix dir…
Skill Claude CodeCodex
Stashes (suspends) the active change — writes a .paused marker so the gates release the single-active-change slot while every artifact (ledger, index, proposal) stays warm for $spec-resume. Refuses on a running $spec-loop change.
Skill Claude CodeCodex
Reports where the current SDD change stands, what artifacts exist, and which commands can be run next. Use when returning after an interruption and unsure where things left off.
Skill Claude CodeCodex
Verifies the current change by dispatching the independent spec-verifier agent (fresh context — the implementing conversation never audits itself) across four dimensions + charter audit; native adds the opt-in project-idiom conformance pass. Every run updates the verification ledger spec/changes/ /verify.md (stable…