viacheslav-tronko/claude-code-harness
Command Claude Code
Surgical fix from RCA — implements §7.1 PRIMARY fix only; NEVER §7.2; NEVER re-analyzes. Triggers: /bug-fix, /fix, "apply this fix", "fix this bug", "implement the fix", RCA-.md asking "fix this". MODIFIES source via Edit/Write — surgical, falsification-first, anti-orthogonal-damage. FM: 2, 5, 12, 16, 18, 19, 20.
viacheslav-tronko/claude-code-harness
Command Claude Code
Full bug-fix pipeline — runs /bug-why → /verify(RCA) → /trace(virtual) → ASK USER → /bug-fix → /trace(real) → /verify(FIX) sequentially. Triggers: /bug-full, "full bug pipeline", "fix this bug end-to-end", "RCA + verify + fix + verify", Jira ticket / error wanting entire pipeline. MODIFIES source at step 5 — Two-Stage…
viacheslav-tronko/claude-code-harness
Command Claude Code
Root cause analysis for software bugs. Triggers: /bug-why, /bug.investigate, "why is this failing", "find the root cause", "trace this exception", "investigate this bug", "post-incident review", stack trace, error log, NullReferenceException, SqlException, 5 Whys, Kepner-Tregoe, IS-IS-NOT, fault localization…
viacheslav-tronko/claude-code-harness
Command Claude Code
Transform vague human requirements (Jira, Confluence, free text, file path) into AI-ready specs — EARS + Given-When-Then + traceable REQ-IDs + glossary links + Open Questions. Triggers: /distill, "analyze requirements", "clarify task", "ambiguous AC", "EARS spec", "GWT", "AI-ready spec", Jira key, Confluence URL, free…
viacheslav-tronko/claude-code-harness
Command Claude Code
Build a falsifiable architectural mental model of a code concept, pattern, term, or component in the legacy codebase. Triggers: /explain, "explain X", "what is X", "how does X work", "trace this concept across layers", "preflight before bug-fix", "build mental model", "F/B/D-XXX pattern", domain term / component name.…
viacheslav-tronko/claude-code-harness
Command Claude Code
Full feature pipeline — runs /distill → /howto-implement → /verify(IMP) → /trace(virtual) → ASK USER → /implement → /trace(real) → /verify(impl) sequentially. Triggers: /feature-full, "full feature pipeline", "spec to code", "implement this story end-to-end", Jira user story / requirement wanting entire pipeline.…
viacheslav-tronko/claude-code-harness
Command Claude Code
Pre-implementation blueprint for AI coding agents on legacy codebases. Triggers: /howto-implement, /blueprint, "implementation plan / analysis / strategy / approach", "how to implement / add / build / refactor", Jira / user story / RCA / SPEC asking "plan this". Investigation-only — produces falsifiable IMP report (3+…
viacheslav-tronko/claude-code-harness
Command Claude Code
Surgical code execution from Implementation Blueprint (IMP) or Plan + TASK-NN. Triggers: /implement, /code, /apply, "implement this", "code this", "build this", "execute the plan", "implement the blueprint", IMP-.md / PLAN-.md asking "implement". MODIFIES source via Edit/Write — surgical-changes-first…
viacheslav-tronko/claude-code-harness
Command Claude Code
Read-only preflight pipeline before bug investigation — runs /explain --for-bug-fix → /bug-why sequentially in main context. Triggers: /preflight, "preflight before bug", "investigate this bug fully", "explain then RCA", Jira ticket / error wanting complete read-only diagnostic. NEVER modifies source. FM: 1, 2, 6, 7 …
viacheslav-tronko/claude-code-harness
Command Claude Code
White-box code-path tracing — REAL mode (empirical CFG/MPC traversal of completed code; Counterfactual + optional Live Mutation Testing; Killer-Path-First; FAIL-FAST) or VIRTUAL mode (pre-implementation falsification of RCA / IMP / Plan). Triggers: /trace, /trace.real, /trace.virtual, /verify-plan, "verify this code"…
viacheslav-tronko/claude-code-harness
Command Claude Code
Hypothesis-driven task-correctness verification before submission. Triggers: /verify, /verify-task, /gate2, "verify this", "verify the RCA", "verify the implementation", "is this fix correct", "double-check my plan", "audit before merge", "pre-flight before PR", "Gate 2 / Pre-Send check", "find what's wrong with my…
viacheslav-tronko/claude-code-harness
Skill Claude CodeCodex
Evidence-bound investigation producing a falsifiable architectural mental model of a code concept, pattern, term, or component in large legacy codebases. Use WHENEVER user asks WHAT/HOW/WHERE about code — /explain, "explain X", "what is X", "how does X work", "trace this concept across layers", "preflight before…
viacheslav-tronko/claude-code-harness
Skill Claude CodeCodex
Pre-implementation blueprint for AI coding agents on legacy codebases. Use WHENEVER user asks HOW to implement, plan, or design — /howto.implement, /blueprint, "implementation plan / analysis / strategy / approach", "how to implement / add / build / refactor", pastes Jira / user story / RCA / SPEC asking "plan this"…
viacheslav-tronko/claude-code-harness
Skill Claude CodeCodex
Evidence-bound transformation of vague human requirements (Jira ticket, Confluence page, free text, file path) into AI-ready specifications for coding agents. Use WHENEVER user asks to /distill, "analyze requirements", "clarify task", "vague requirements", "ambiguous AC", "task vs bug", "EARS", "Given-When-Then"…
viacheslav-tronko/claude-code-harness
Skill Claude CodeCodex
Root cause analysis for software bugs in large legacy codebases. Use this skill WHENEVER the user asks WHY a bug happens, asks for RCA, mentions root cause analysis, says /bug.why, /bug.investigate, asks "why is this failing", "find the root cause", "trace this exception", "investigate this bug", "post-incident…
viacheslav-tronko/claude-code-harness
Skill Claude CodeCodex
Surgical code-execution skill for AI coding agents on legacy codebases. Use WHENEVER the user asks to IMPLEMENT, CODE, BUILD, APPLY, or FIX from a contract source — IMP / Plan + TASK-NN, RCA report, freeform task, or Jira ticket. Triggers on /implement, /bug.fix, /code, /apply, /fix, "implement this", "code this"…
viacheslav-tronko/claude-code-harness
Skill Claude CodeCodex
Hypothesis-driven task-correctness verification for AI coding agents on legacy codebases. Use WHENEVER user asks to verify, audit, or pre-flight a task artifact (IMP, RCA, Plan, Implementation, Fix, Jira ticket) before submission — /verify, /verify-task, /gate2, "verify this", "verify the RCA", "verify the…
viacheslav-tronko/claude-code-harness
Skill Claude CodeCodex
White-box code-path tracing in two modes. REAL — empirical verification of completed code via CFG/MPC traversal, Counterfactual + optional Live Mutation Testing, PBT hook, Tandem-FL on FAIL, Killer-Path-First, FAIL-FAST, persistent index-prefixed JSONL trace with resume + confidence decay, Reviewer Agent re-rank…
viacheslav-tronko/claude-code-harness
Instructions file
Instructions for viacheslav-tronko/claude-code-harness, covering claude.md — operational doctrine ({yourprojectname}), projectconfig — fill these placeholders before using, 0. top of mind — these four ride above everything, 1. mission and 2. p0 hard constraints — never violate.