MichelKerkmeester /skilled-agent-harness_spec-driven-loops
Skill Claude Code OpenCode
Domain evidence and shared workflow doctrine for browser/Webflow frontend work. This surface owns the implement -> debug -> verify phases through the workflow references below, and acts on the evidence gathered here when the hub detects a Webflow surface. Detection markers: src/2javascript, webflow, --vw- custom…
★ not rated 35 ▲
+1
changed yesterday A SkillSpector: pass 44 tokens
original MIT MichelKerkmeester /skilled-agent-harness_spec-driven-loops
Skill Claude Code OpenCode
Projects terse CLI output to plain English byte-safely, and explains a topic or reply as the smallest visual at a chosen depth.
★ not rated 35 ▲
+1 yesterday A SkillSpector: pass 31 tokens
original MIT MichelKerkmeester /skilled-agent-harness_spec-driven-loops
Skill Claude Code OpenCode needs its repo
Extracts a live website's real CSS into a v3 Style Reference DESIGN.md via an embedded extract-write-validate pipeline.
★ not rated 35 ▲
+1
changed yesterday C 30 tokens
original MIT MichelKerkmeester /skilled-agent-harness_spec-driven-loops
Skill Claude Code OpenCode
Design parent hub. Routes one design identity to the mode that owns the decision being asked for, starting with sk-design-fundamentals.
★ not rated 35 ▲
+1
changed 3d ago A SkillSpector: pass 31 tokens
original MIT MichelKerkmeester /skilled-agent-harness_spec-driven-loops
Skill Claude Code OpenCode needs its repo
Documentation & OpenCode-component authoring hub: skills, agents, commands, READMEs, catalogs, playbooks, changelogs, frontmatter.
★ not rated 35 ▲
+1
changed yesterday A SkillSpector: pass 34 tokens
original MIT MichelKerkmeester /skilled-agent-harness_spec-driven-loops
Skill Claude Code OpenCode needs its repo
Scaffold OpenCode agents with runtime placement, permission frontmatter, authority boundaries, workflow sections, and validation.
★ not rated 35 ▲
+1
changed 6d ago A SkillSpector: pass 26 tokens
original MIT MichelKerkmeester /skilled-agent-harness_spec-driven-loops
Skill Claude Code OpenCode needs its repo
Author MCP-promotion, behavior, skill-benchmark, and model-benchmark artifacts; route the Lane A authoring guide.
★ not rated 35 ▲
+1 yesterday A SkillSpector: pass 30 tokens
original MIT MichelKerkmeester /skilled-agent-harness_spec-driven-loops
Skill Claude Code OpenCode needs its repo
Author global or packet-local changelogs with topology detection, versions, and release notes.
★ not rated 35 ▲
+1
changed 6d ago A SkillSpector: pass 23 tokens
original MIT MichelKerkmeester /skilled-agent-harness_spec-driven-loops
Skill Claude Code OpenCode needs its repo
Scaffold OpenCode slash commands with explicit argument hints, allowed tools, and router/presentation separation.
★ not rated 35 ▲
+1
changed 6d ago A SkillSpector: warn 24 tokens
original MIT MichelKerkmeester /skilled-agent-harness_spec-driven-loops
Skill Claude Code OpenCode
Local, Git-free before/after review of an edited document (text, Markdown, HTML, DOCX, text PDF) as a self-contained HTML report.
★ not rated 35 ▲
+1 yesterday A SkillSpector: pass 37 tokens
original MIT MichelKerkmeester /skilled-agent-harness_spec-driven-loops
Skill Claude Code OpenCode needs its repo
Create sk-doc feature-catalog packages with a root catalog, category folders, per-feature files, and auditable source anchors.
★ not rated 35 ▲
+1 yesterday A SkillSpector: warn 31 tokens
original MIT MichelKerkmeester /skilled-agent-harness_spec-driven-loops
Skill Claude Code OpenCode
Own the YAML frontmatter contract: which fields a document class carries, the description budget, and the 4-part version field.
★ not rated 35 ▲
+1
changed 4d ago A SkillSpector: warn 0 tokens
original MIT MichelKerkmeester /skilled-agent-harness_spec-driven-loops
Skill Claude Code OpenCode needs its repo
Author manual testing playbook packages with deterministic scenarios, evidence collection, and multi-agent execution planning.
★ not rated 35 ▲
+1
changed 6d ago A SkillSpector: warn 27 tokens
original MIT MichelKerkmeester /skilled-agent-harness_spec-driven-loops
Skill Claude Code OpenCode needs its repo
Validate, score, and optionally improve existing markdown via structure extraction, DQI scoring, HVR review, and validation gates.
★ not rated 35 ▲
+1 yesterday A SkillSpector: pass 30 tokens
original MIT MichelKerkmeester /skilled-agent-harness_spec-driven-loops
Skill Claude Code OpenCode needs its repo
Author sk-doc folder, code-folder and skill/project READMEs plus folded five-phase install guides.
★ not rated 35 ▲
+1 yesterday A SkillSpector: warn 25 tokens
original MIT MichelKerkmeester /skilled-agent-harness_spec-driven-loops
Skill Claude Code OpenCode
Author, revise or retire a repo rule from a user's request, with the decision tests that refuse most of them.
★ not rated 35 ▲
+1
changed 6d ago A SkillSpector: pass 29 tokens
original MIT MichelKerkmeester /skilled-agent-harness_spec-driven-loops
Skill Claude Code OpenCode needs its repo
Scaffold OpenCode skills and two-axis sk-doc parent hubs, including standalone, nested workflow, and surface packets.
★ not rated 35 ▲
+1 yesterday A SkillSpector: warn 28 tokens
original MIT MichelKerkmeester /skilled-agent-harness_spec-driven-loops
Skill Claude Code OpenCode
Apply the Human Voice Rules to prose, or score prose against them, with a scope gate first and a re-scan after the rewrite.
★ not rated 35 ▲
+1
changed 6d ago A SkillSpector: pass 35 tokens
original MIT MichelKerkmeester /skilled-agent-harness_spec-driven-loops
Skill Claude Code OpenCode
Git: numbered worktrees, conventional commits, PRs, merge/rebase, and finish; single-skill workflow guidance with no spec.
★ not rated 35 ▲
+1
changed 6d ago A ✓ AI review 32 tokens
original MIT MichelKerkmeester /skilled-agent-harness_spec-driven-loops
Skill Claude Code OpenCode
Prompt engineering: transforms a request into a structured, scored AI prompt via 7 frameworks, DEPTH thinking and CLEAR scoring.
★ not rated 35 ▲
+1
changed 6d ago A SkillSpector: warn 29 tokens
original MIT MichelKerkmeester /skilled-agent-harness_spec-driven-loops
Skill Claude Code OpenCode
Local vision for text-only models: OCR, inspect, detect and pixel analysis on screenshots via a private Moondream runtime.
★ not rated 35 ▲
+1 yesterday A SkillSpector: warn 29 tokens
original MIT MichelKerkmeester /skilled-agent-harness_spec-driven-loops
Skill Claude Code OpenCode needs its repo
Routes research, review, AI Council, and improvement modes through registry-selected packets.
★ not rated 35 ▲
+1
changed 6d ago C 21 tokens
original MIT MichelKerkmeester /skilled-agent-harness_spec-driven-loops
Skill Claude Code OpenCode needs its repo
AI Council: multi-seat planning, artifact persistence, convergence checks, packet-local ai-council outputs.
★ not rated 35 ▲
+1
changed yesterday A SkillSpector: pass 27 tokens
original MIT MichelKerkmeester /skilled-agent-harness_spec-driven-loops
Skill Claude Code OpenCode
Evaluator-first bounded agent improvement: 5-dim scoring, dynamic profiling, packet-local candidates, guarded promotion.
★ not rated 35 ▲
+1
changed yesterday A SkillSpector: pass 26 tokens
original MIT 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: