cwijayasundara/claude_harness_eng_v5

code-gen

Code generation quality principles — TDD, typing, error handling, logging, API integration, LLM integration.

Langue du texte source : anglais

author-prompt-surface

Use when creating or editing .claude/agents/*.md, .claude/skills/*/SKILL.md, or .claude/commands/* — apply docs/prompting-standards.md (trigger conditions, model-agnostic bodies, no reasoning-as-text, effort floors).

Langue du texte source : anglais

auto

Autonomous build loop with Karpathy ratcheting, GAN evaluator, and session chaining. Iterates story groups until all features pass or stopping criteria met.

Langue du texte source : anglais

retro

Review this harness's own recent loop-health signals and recommend scored, evidence-backed improvements to the harness itself (skills, gates, sensors, CLAUDE.md rules) — never to product code. Use after a `/auto`/`/gate` run, on a drift cadence, or on demand…

Langue du texte source : anglais

attestation

Generate a per-commit, durable compliance attestation (with a sha256 corruption-detecting integrity checksum) from the harness control inventory, the branch-protection/deploy-gate verify outputs, the gate verdict, and the ratchet baselines — then commit the…

Langue du texte source : anglais

portfolio-rollup

Aggregate per-repo compliance attestations into one integrity-hashed portfolio rollup — verifying each attestation's sha256 integrity, recording an unattested repo as a compliance gap, and computing harness-version drift across the fleet. Use over a local…

Langue du texte source : anglais

provision

Provision the GitHub security-compliance gates and verify they hold. Three modes — protection (branch-protection ruleset so gitleaks+sast+code-owner review block merges), environments (deploy-approval environment so a deploy pauses for a required reviewer),…

Langue du texte source : anglais

change

Change the behavior of existing code — story-driven by default, or --issue N for a GitHub bug fix. Test-first, full verification, code review.

Langue du texte source : anglais

vibe

Controlled small-change lane for low-risk fixes, docs, tests, and narrowly scoped edits without running the full SDLC pipeline.

Langue du texte source : anglais

gate

Run the adaptive pre-merge quality gate: deterministic checks, evaluator, diff review, and security review only when the diff crosses a security/data/API boundary. (Renamed from /review to avoid colliding with Claude Code's native /review PR-review command.)

Langue du texte source : anglais

design

[Internal pipeline stage — run by /build (use --doc-only standalone for an ARB narrative); invoke directly only as a power user.] Generate system architecture, machine-readable schemas, and UI mockups. Spawns planner + generator concurrently.

Langue du texte source : anglais

brd

[Internal pipeline stage — run by /build; invoke directly only as a power user.] Create a Business Requirements Document — from a Socratic interview, or grounded in a Functional Requirements Document via --frd (with a deterministic net-new/dropped gate).…

Langue du texte source : anglais

spec

[Internal pipeline stage — run by /build; invoke directly only as a power user.] Decompose BRD into epics, stories, dependency graph, and feature list for agent team execution.

Langue du texte source : anglais

code-map

[Internal pipeline stage — run by /brownfield and /seam-finder; invoke directly only as a power user.] Build a deterministic dependency graph of an existing codebase. AST-first for Python, React/JS/TS, and Java/C#/Go (stdlib ast + tree-sitter wheels — symbols…

Langue du texte source : anglais

feature

Brownfield change route — take an existing-code feature request from intent to a reviewed PR, scaling from a single /change to an epic via /spec→/design→/auto. Linear-tracked, backed by a committed DeepWiki.

Langue du texte source : anglais

reuse-or-justify

Use when reuse-scout fires at intake (band ≥ medium, a touched constitution invariant, or a same-release clone cluster) — run the confidence-gated reuse-vs-new dialogue and record the decision + performance budget before code. [Internal discipline — invoked…

Langue du texte source : anglais

sprint

PRD-per-sprint evolution route for an existing harness-built (or brownfield) system — grounds a new PRD against the prior requirement spine, produces a human-approved design amendment against the living specs/design/ baseline, then runs /auto. Companion to…

Langue du texte source : anglais

test

[Internal pipeline stage — run by /build and /auto; invoke directly only as a power user.] Generate test plan, test cases, test data fixtures, and Playwright E2E tests mapped to acceptance criteria.

Langue du texte source : anglais

writing-acceptance-tests-first

Use when a story is about to move from acceptance criteria to implementation — in /test's plan-phase deliverable or /change Step S4, before any production code is written — to write a business-readable acceptance test against a Ports-and-Adapters seam with a…

Langue du texte source : anglais

evaluate

[Internal pipeline stage — run by /auto and /gate; invoke directly only as a power user.] Run the application and verify sprint contract criteria via API tests, Playwright interaction, and schema validation.

Langue du texte source : anglais

promote

Implement an approved /retro recommendation and open a PR against this harness repo. Use after a human has approved a recommendation via /retro --apply-decisions. Merge always stays human — this skill only opens the PR, never merges. Refuses…

Langue du texte source : anglais

fix-from-diagnostics

[Internal pipeline stage — run by /auto self-heal and /implement validation when lint/type error volume is high; invoke directly only as a power user.] Turn compiler/linter output into a sharded work queue; fix per package with optional adversarial review —…

Langue du texte source : anglais

implement

[Internal pipeline stage — run by /auto (follow with /gate); invoke directly only as a power user.] Generate production code and tests for a story group using agent teams for parallel execution.

Langue du texte source : anglais

refactor

Refactor existing code for quality, performance, or maintainability. Enforces core quality principles with ratchet gate.

Langue du texte source : anglais

build

Full SDLC pipeline. Runs all phases end-to-end with human gates on phases 1-3.

Langue du texte source : anglais

advise

Invoke the frontier advisor for a structural re-rank / unblock / design pivot. Compact brief only.

Langue du texte source : anglais

brownfield

Discover and map an existing codebase before planning or changing it. Add --seams "<goal>" to also rank the safest cut-points for that goal.

Langue du texte source : anglais

context

Build a bounded, cited context pack from the living DeepWiki/code-map before broad source reads. Use in brownfield lanes and whenever source orientation could otherwise require large file reads.

Langue du texte source : anglais

scaffold-upgrade

Upgrade harness control-plane files (hooks, scripts, git-hooks, agents) in a previously scaffolded project without wiping project-manifest or state. Dry-run by default.

Langue du texte source : anglais

upgrading-dependencies

Use when bumping a dependency version — package.json, pyproject.toml, requirements, or lockfile changes — in an existing codebase. Classifies the bump, audits the usage surface from the code graph, and isolates the upgrade in its own proven commit.

Langue du texte source : anglais

sprouting-instead-of-editing

Use when a symbol that must change is UNCOVERED and unpinnable (seam score < 0.5, or a skeletons/-flagged god file with no clean boundary) — adds behavior in a new fully-tested unit instead of editing legacy code in place. [Internal discipline — applied…

Langue du texte source : anglais

agent-readiness

Show how mature/safe this codebase is for heavy AI-agent use — a synthesis dashboard across 8 pillars, aggregating signals the harness already collects. Read-only report; aggregates state already on disk.

Langue du texte source : anglais

checking-coverage-before-change

Use when about to edit any existing (non-greenfield) symbol — before the first Edit/Write to production code in /refactor, /change, or /vibe — to learn which tests cover it and route uncovered code to pinning or sprouting. [Internal discipline — applied…

Langue du texte source : anglais

pinning-down-behavior

Use when checking-coverage-before-change reports an UNCOVERED symbol you must edit — writes characterization (pin-down) tests at the nearest observable seam and verifies they bite before any production edit. [Internal discipline — applied automatically by…

Langue du texte source : anglais

install-framework-packs

[Internal pipeline stage — run by /scaffold; invoke directly only as a power user.] Verify agent-framework skill packs declared in project-manifest.json#framework_skill_packs and print normal-terminal install commands for missing packs. Use after /scaffold…

Langue du texte source : anglais

react-code

Build client-side React apps with Vite — useEffect cleanup and stale-closure bugs, dependency-array gotchas, Context vs. server-state libraries for API data, Vite environment variables, and testing with Vitest. Scoped to Vite+React client-side apps — NOT…

Langue du texte source : anglais

fastapi-code

Build FastAPI backends — dependency injection with Depends(), Pydantic v2 request/response validation, async vs sync route rules, background tasks, and testing with TestClient. Use for any FastAPI route, dependency, or Pydantic model. Not for database/ORM…

Langue du texte source : anglais

deepagents-code

Build long-running, production-ready agents with DeepAgents — the opinionated harness on top of LangGraph/LangChain providing filesystem access, task planning, subagent delegation, and context management out of the box. Use for long-running…

Langue du texte source : anglais

langchain-code

Build LangChain agents in Python — create_agent, chat model initialization, tool binding, structured output, and middleware. Use for a single-agent tool-calling loop with a system prompt and a fixed tool list; use langgraph-code instead when the task needs…

Langue du texte source : anglais

langgraph-code

Build stateful, multi-actor LangGraph agents in Python — StateGraph definition, conditional routing, checkpointing/persistence, and human-in-the-loop interrupts. Use when a task needs explicit multi-step control flow, crash/interrupt recovery, or approval…

Langue du texte source : anglais