Build
Run a single-thread plan-and-execute loop against SPEC.md tasks, updating status cells and invoking backprop when verification fails.
Overview
build is an agent skill for the Build phase that plan-executes SPEC.md §T tasks in a single thread and invokes backprop when verification fails.
Install
npx skills add https://github.com/juliusbrussee/blueprint --skill buildWhat is this skill?
- Native plan→execute loop with no sub-agents; main Claude owns the whole run
- Loads SPEC.md and FORMAT.md; stops and defers to spec skill if SPEC.md is missing
- Task targeting via §T.n, --next (lowest open row), or --all for every pending dot row
- Plan step cites §V invariants and §I interfaces before edits; user OK unless auto mode
- On verification fail, auto-invokes backprop before retry to capture new §V invariants
Adoption & trust: 5 installs on skills.sh; 1k GitHub stars; 3/3 security scanners passed (skills.sh audits).
What problem does it solve?
You have SPEC.md tasks but no disciplined loop that respects invariants, updates task status, and learns from failed verifications.
Who is it for?
Blueprint users with SPEC.md and FORMAT.md who want --next or §T-targeted implementation without multi-agent orchestration.
Skip if: Greenfield repos with no SPEC.md—invoke the spec skill first instead of improvising build.
When should I use this skill?
Triggers when the user asks to build, implement, execute the spec, or tackle a specific §T task (`build §T.3`, `build --next`, `implement next task`, `run the build`). Expects SPEC.md to exist.
What do I get? / Deliverables
Chosen §T tasks move to done with passing verification, or failures trigger backprop before retry so §V invariants capture what broke.
- Code edits per §T plan
- Updated §T status cells in SPEC.md
- Passing verification command or backprop-driven invariant updates
Recommended Skills
Journey fit
How it compares
Spec-task executor with invariant gates, not a free-form 'just implement my app' codegen prompt.
Common Questions / FAQ
Who is build for?
Solo builders using Julius Brussee's Blueprint spec format who want one agent thread to execute §T tasks with verification.
When should I use build?
In Build when you say build, implement the spec, build --next, build §T.n, or run the build after SPEC.md exists.
Is build safe to install?
It edits code and runs tests or linters; review the Security Audits panel on this page before allowing shell and filesystem access.
SKILL.md
READMESKILL.md - Build
# build — implement spec Single-thread native plan→execute. You are main Claude. No swarm. ## LOAD 1. Read `SPEC.md`. If missing → tell user to invoke the spec skill first. Stop. 2. Read `FORMAT.md` once if not loaded. 3. Parse invocation args: - `§T.n` → that task only - `--next` → lowest-numbered row with status `.` or `~` - `--all` or empty → every `.` row in §T order ## PLAN Native plan mode. For chosen task(s): 1. Cite every §V invariant that applies. Plan must respect all. 2. Cite every §I interface touched. Plan must preserve shape. 3. List files to create / edit. 4. List tests to add or update (one per invariant touched). 5. Name verification command (test, build, lint). Show plan. Wait for user OK unless auto mode. ## EXECUTE Per task in order: 1. Flip §T.n status cell `.` → `~`. Just write to SPEC.md. 2. Edit code per plan. 3. Run verification command. 4. **Pass** → flip `~` → `x`. Next task. 5. **Fail** → invoke backprop skill. Do NOT retry blindly. ## FAIL → BACKPROP On test/build failure: 1. Read failure output. 2. Ask: is failure (a) my code bug, (b) spec wrong, or (c) unspecified edge case? 3. If (a) → fix code, re-run. No spec change. 4. If (b) or (c) → invoke spec skill with `bug: <cause>` first, let it update §V and §B, then resume build against updated spec. Rule: never silently fix root-cause without considering backprop. §B is the memory that stops recurrence. ## WRITE POLICY - Only flip §T status. No other SPEC.md edits from build. - Other spec edits → invoke spec skill. - Commit after each §T completes. Message: `T<n>: <goal line>` + §V cites. ## VERIFICATION Task `x` only if: - Verification command exits 0. - New test(s) added per plan. - No §V invariant regressed (run full test suite at end). ## NON-GOALS - No sub-agents. No parallel workers. Main thread only. - No progress dashboards. `cat SPEC.md | grep §T` is the dashboard. - No speculative work beyond chosen task scope.