geniro-io

geniro-io/geniro-claude-harness

Plugin Claude Code

Plugin marketplace listing 1 plugin: geniro.

geniro-io/geniro-claude-harness

Plugin Claude Code

Production-grade Claude Code harness with AI-driven setup, multi-agent skills, and safety hooks. Includes /setup for project-tailored configuration.

geniro-io/geniro-claude-harness

Skill Claude CodeCodex

Use when post-hoc analyzing saved Claude conversation threads for pipeline errors — wrong subagent tier, missed parallel-spawn, schema-invalid tool args, premature completion, instruction drift, hallucinated tools. Also audits coverage: whether every custom-instruction file and block (rules, constraints, additional…

geniro-io/geniro-claude-harness

Skill Claude CodeCodex

Use when auditing the Geniro plugin repo as a whole — skills, agents, hooks, lib helpers, rules, and docs — for cross-file consistency, stale references, authoring-rule compliance, logic and shell correctness, over-complication, magic numbers, safety/coverage gaps, and wiring completeness (what the plugin declares but…

geniro-io/geniro-claude-harness

Skill Claude CodeCodex

Use when improving a plugin module through the evals/loop measurement cycle — proposing a hypothesis, screening a variant against the champion, judging runs, reading verdicts, confirming on holdout, promoting a winner, calibrating the judge, or growing a benchmark. Owns the judgment half of the loop (error analysis…

geniro-io/geniro-claude-harness

Skill Claude CodeCodex

Use when finding past Claude Code conversation threads — across every project and every config dir on this machine — to hand selected ones to /analyze-thread for failure analysis. Scans the projects/ tree of each config dir, keeps threads that did agentic work (edited code, ran a skill, or spawned a subagent — so…

geniro-io/geniro-claude-harness

Skill Claude CodeCodex

Use when modifying the Geniro plugin itself — fix a Geniro skill, agent, hook, or ARCHITECTURE.md. Researches via parallel agents (codebase + ARCHITECTURE.md + internet), presents evidence, implements after approval. Skip for general codebase Q&A (/geniro:investigate) or app-code bugs (/geniro:debug).

geniro-io/geniro-claude-harness

Instructions file

Instructions for geniro-io/geniro-claude-harness, covering geniro plugin, skill routing, path rules, never force-add .geniro/ paths and state files.

geniro-io/geniro-claude-harness

Agent

Read-only codebase reconnaissance. Use at Phase 1 of an implementation task to scope a spec.md (or inline task) — identifies likely-touched files, 2-3 exemplar files to mirror, matching .claude/rules/ entries, a REUSE-AS-IS / EXTEND / NO-ANALOGUE inventory, risk-signal flags, and a change-scope estimate (trivial /…

geniro-io/geniro-claude-harness

Agent

Read-only general codebase research. Use when a skill needs to map a subsystem, trace a flow, locate a definition, or summarise behaviour across files — anywhere a multi-file investigation would otherwise flood the orchestrator's context with file contents. Returns a structured findings table with file:line citations…

geniro-io/geniro-claude-harness

Agent

Independent verifier for an already-raised finding. Use when a run needs a second, uncontaminated judgment on a specific claim — /geniro:review Phase 4.2 per-finding verification, /geniro:resolve verdict re-verification, a spec-claim challenge, or a user's Challenge-this-finding pick. Re-reads the cited code cold…

geniro-io/geniro-claude-harness

Agent

Read-only past-knowledge search across the memory layers. Use at /geniro:implement Phase 1 for a full multi-layer sweep — past learnings + project snapshots + prior review/debug handoffs + prior plans. /geniro:review, /geniro:debug, /geniro:refactor spawn it scoped to just the backend learnings read (SCOPE…

geniro-io/geniro-claude-harness

Agent

Post-task improvement synthesizer. Spawned by /geniro:reflect (user-invoked, on-demand) to extract durable project-rule candidates from recent work — a diff, a finding set, or session extracts drawn from past transcripts or from the running session — routed to CLAUDE.md / .claude/rules/ / .geniro/instructions/ / ADR /…

geniro-io/geniro-claude-harness

Agent

Single-dimension code reviewer. Use when /geniro:review Phase 2 or /geniro:implement Phase 3 self-review spawns parallel reviewers — one instance per dimension (bugs / security / architecture / tests / optimizations / conventions / regressions / design / pr-metadata / spec-compliance / code-quality). Returns…

geniro-io/geniro-claude-harness

Agent

Executes the project's pre-resolved TESTCOMMAND once and returns a structured pass/fail summary with up to 15 failure snippets. Use at end-of-phase test runs and inside fix-retry loops so the raw test stdout (typically 50K+ tokens) never reaches the orchestrator's main context.

geniro-io/geniro-claude-harness

Skill Claude CodeCodex

Use when scaffolding a reusable workflow-helper (Slack/PR/release automations) or invoking a previously-created action. Stored at .geniro/actions/. Run-mode executes the action directly — invoking it is the authorization, so no confirmation is asked. Skip for editing core Geniro skills — edit the plugin repo directly.

geniro-io/geniro-claude-harness

Skill Claude CodeCodex

Use when auditing the current repo's AI-assistant instruction files — CLAUDE.md, AGENTS.md, Cursor rules, Copilot instructions, and every other agent-facing surface — for accuracy against the codebase, cross-tool consistency, bloat, structure, and coverage gaps: 'audit our AI instructions', 'is CLAUDE.md stale', 'are…

geniro-io/geniro-claude-harness

Skill Claude CodeCodex

Use when a bug needs systematic investigation. 3-phase loop (Investigate → Propose → Ship) mirroring /geniro:implement: observe → hypothesize → test → isolate → propose fix → author reproduction test, then escalate to /geniro:implement with a handoff file at .geniro/state/handoff/from-debug- .md. Adversarial mode…

geniro-io/geniro-claude-harness

Skill Claude CodeCodex

Use when shipping a new feature, endpoint, page, or significant change against a spec.md / plan.md (from /geniro:plan) OR a raw inline task description. 3-phase autonomous loop: Analyze → Implement → Self-review-and-Ship. Optional --deep deepens two phases — a multi-angle self-review with verification escalated only…

geniro-io/geniro-claude-harness

Skill Claude CodeCodex

Use when adding skill-behavior rules at Geniro skill phase boundaries OR cross-cutting code-style rules loaded at every code-writing/review step; also for declaring read-only fact-verification sources (## Data Sources), recording what each project check covers and leaves uncovered so results are not overstated (##…

geniro-io/geniro-claude-harness

Skill Claude CodeCodex

Use when answering deep codebase questions that need evidence — repo structure, code behavior, git history, or internet sources. Parallel research agents produce cited answers. Skip for bug fixes (/geniro:debug) or codebase mapping (/geniro:onboard).

geniro-io/geniro-claude-harness

Skill Claude CodeCodex

Use when starting fresh in an unfamiliar codebase and need rapid orientation. Scans structure and conventions; produces CODEBASEMAP.md with architecture, module graph, critical paths, entry points. Skip for specific Q&A (/geniro:investigate) or bug investigation (/geniro:debug).

geniro-io/geniro-claude-harness

Skill Claude CodeCodex

Use when turning a vague idea or feature request into an approved spec.md before /geniro:implement. Spec-first planning workflow: explore → grill (decision-tree clarification) → propose 2-3 approaches → approve sections → write spec.md → mechanical validate → user approve → handoff. Skip for well-formed specs already…

geniro-io/geniro-claude-harness

Skill Claude CodeCodex

Use when restructuring code for better organization or reducing tech debt with zero behavior change. 3-phase loop (Plan → Apply → Verify); never ships — the diff is the deliverable. For behavioral changes use /geniro:implement; for performance use /geniro:review (optimizations dimension).