skills/blueprint-specialist/SKILL.md
Blueprint reconstruction skill for rebuilding a bounded deterministic slice when runnable source is missing; use when only pseudo-code, plans, or artifact evidence remain and a truthful re-implementation with proof is required.
npx skillsauth add ic1558/0luka blueprint-specialistInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Use this skill when runnable source code is missing or unusable, but enough grounded blueprint material exists to re-implement a bounded executable slice safely.
This skill is for truthful reconstruction from:
This skill is not for pretending recovery happened when source is gone.
Produce a bounded, executable, non-placeholder rebuild from grounded blueprint materials.
The only valid end states are:
Before using this skill for any governed 0luka lane:
/Users/icmini/0luka/core_brain/governance/AGENTS.yaml/Users/icmini/0luka/CONTRACT.yaml/Users/icmini/0luka/skills/manifest.mdDo not begin execution planning before these are read.
Use this skill when one or more are true:
Do not use this skill when:
If runnable source is gone and the implementation is recreated from notes, pseudo-code, plans, or artifacts:
Recovery is only valid if real runnable source is actually restored.
At least one grounded source must exist:
If none exist, return:
FAIL / blueprint insufficientThink like a forensic rebuilder.
That means:
Verify all of the following:
Return exact evidence for each.
List all grounded reconstruction inputs.
Classify each as:
primary_blueprintsecondary_supportartifact_evidencecontext_onlyPrefer direct pseudo-code over loose surrounding notes.
Choose the smallest slice that proves the system is no longer placeholder-only.
This slice must be:
Do not attempt whole-system rebuild unless the blueprint is unusually complete and the lane explicitly allows it.
For each intended target file, state:
This mapping step is mandatory.
Write real bounded code only for the chosen slice.
Rules:
Proof must show actual input -> output behavior.
At minimum prove:
Be explicit about:
Do not imply full restoration if only a bounded slice was rebuilt.
Prefer blueprint sources in this order:
If sources conflict:
If the original intent is deterministic, the rebuild must preserve that.
For deterministic rebuilds:
If uncertainty exists:
Allowed:
Not allowed:
Return only:
This skill is done only when:
Pseudo-code exists, source is gone, minimum engine slice is rebuilt and proven with explicit inputs/outputs.
Old outputs exist, formulas survive in notes, rebuild reproduces the bounded expected behavior.
Core slice is rebuilt, but scanner/runner/integration layers remain unrebuilt and are explicitly listed as next-lane work.
Notes are too vague, no trustworthy formulas remain, so the skill returns FAIL instead of hallucinating a rebuild.
Do not ask:
How do I recreate the whole missing system?
Ask:
What is the smallest truthful executable slice I can rebuild from the blueprint and prove now?
End of Document
development
--- name: auto-workflow-creator description: Create a new skill from a workflow you just performed. Generates a valid 3-layer Anthropic-style skill — YAML frontmatter (name + description + triggers), imperative instruction body, and optional scripts/ subdirectory — and prints the registration commands per target runtime (Claude Code auto-discovers from ~/.claude/skills/; Codex needs an explicit [[skills.config]] entry in ~/.codex/config.toml; OpenCode needs a registry entry in ~/.opencode/openco
tools
# Cost-Control Skill Enforce token discipline and model routing optimization (Flash vs Pro) via the Gemini CLI 0.40 Gemma local router for governed-agent workflows. ## Guidelines - **Preference for Local/RTK/File Tools:** Always prioritize local tools (ls, read_file, grep_search) and RTK for initial data gathering before calling model logic. - **Flash-First Routing:** - Route all routine code analysis, file reads, summarization, search, and classification tasks to `gemini-3.5-flash-preview` o
development
--- name: verify-first description: Run verification gates before and after code changes, prioritize health and regressions, and fail closed on missing evidence. Mandatory Read: YES --- # Verify First Mandatory Read: YES ## Workflow 1. Capture baseline verify results. 2. Apply minimal changes. 3. Re-run targeted and system-level verify commands. 4. Report explicit pass/fail evidence. ## Caps - Run deterministic verification commands. - Compare before/after outcomes. - Block claims without pr
testing
Use this skill when you need to perform a READ-ONLY forensic validation of a completed task (evidence), verify compliance against a TaskSpec, or issue a formal verdict (PASS/FAIL) before GMX approval. Strictly forbids modification or execution.