Multi-model agent implementation workflow skill for Claude Code. Orchestrates research, evaluation, design baseline, implementation, RCA, and auditing across AI models.
nestharus/agent-implementation-skill
Agent
Merges multiple research ticket results into a cohesive dossier, produces research-derived surfaces in the existing surfaces schema, and writes a proposal addendum for the integration proposer.
nestharus/agent-implementation-skill
Agent
Verifies research dossier claims against their cited sources, flags unsupported or hallucinated claims, and suggests re-scrapes for stale citations.
nestharus/agent-implementation-skill
Agent
Converts a ROAL risk assessment into the minimum effective execution posture and per-step execution decisions.
nestharus/agent-implementation-skill
Agent
Assesses execution risk for a scoped ROAL package before an execution agent starts mutating code or coordinating work.
nestharus/agent-implementation-skill
Agent
Evaluates technical stack alternatives against verified governance, producing comparative risk profiles.
nestharus/agent-implementation-skill
Agent
Explores a codebase heuristically and builds a routing map (codemap) that enables downstream agents to find the right files without exhaustive search.
nestharus/agent-implementation-skill
Agent
Evaluates whether an existing codemap is still valid given codespace changes, producing a rebuild-or-keep decision as structured JSON.
nestharus/agent-implementation-skill
Agent
Performs a fast, coarse-grained exploration of a codebase to identify top-level modules and subsystems without drilling into internals. Produces a skeleton codemap suitable for guiding per-module deep exploration.
nestharus/agent-implementation-skill
Agent
Combines module-level exploration fragments and a skeleton codemap into a unified codemap with a complete routing table.
nestharus/agent-implementation-skill
Agent
Spot-checks a codemap's routing claims against actual codespace state, producing corrections for any inaccuracies found.
nestharus/agent-implementation-skill
Agent
Performs deep analysis of a specific file's relevance to a section, producing structured findings about what matters and what was missed.
nestharus/agent-implementation-skill
Agent
Explores a single module or subsystem in depth, producing a detailed routing fragment that covers internal structure, key files, interfaces, and relationships.
nestharus/agent-implementation-skill
Agent
Compares a section's current related-files list against evidence and emits a structured update signal with additions and removals.
nestharus/agent-implementation-skill
Agent
Uses a codemap to identify which files in the codebase are relevant to a specific section's concerns, producing a targeted related-files list with reasoning.
nestharus/agent-implementation-skill
Agent
Ranks candidate files by relevance tier for deep analysis, deciding which tiers warrant immediate scanning based on section complexity.
nestharus/agent-implementation-skill
Agent
Strategic agent that reads all substrate shards, identifies convergence patterns and contradictions, prunes to a minimal set of shared seams, and produces the substrate artifact and seed plan. This is graph exploration with pruning — not architecture invention.
nestharus/agent-implementation-skill
Agent
Creates minimal anchor files in codespace from the seed plan, and emits related-files update signals so section specs gain references to the new anchors.
nestharus/agent-implementation-skill
Agent
Produces structured shard JSON describing what a section needs, provides, and what shared seams it touches. Explores the section spec and intent pack to surface cross-section dependencies without proposing architecture.
nestharus/agent-implementation-skill
Agent
Decomposes an over-complex section into 2-4 child sections with explicit scope grants and writes child section specs plus a JSON manifest.
nestharus/agent-implementation-skill
Agent
Analyzes cross-section impact of changes. Reads modified files and consequence context to determine which other sections are materially affected and why.
nestharus/agent-implementation-skill
Agent
Extracts structured impact JSON from raw, malformed agent output. Reads a raw text file and produces well-formed impact entries.
nestharus/agent-implementation-skill
Agent
Checks shape and direction of integration proposals and implementations. Verifies alignment between adjacent layers — not tiny details. Returns ALIGNED, PROBLEMS, or UNDERSPECIFIED.
nestharus/agent-implementation-skill
Agent
Classifies alignment check output into a structured verdict when the primary alignment check did not return parseable JSON. Reads raw text output and produces {aligned, frameok, problems}.
nestharus/agent-implementation-skill
Agent
Lightweight classifier that reads agent output and determines its state. Used when output is ambiguous and simple pattern matching cannot reliably classify it. Returns structured JSON.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: