harness-dev

Plugin harness-dev in popoffvg/dotfiles. Commands read from the repo's manifest — 4 days ago.

Skills (8)

• agent-metrics

  Report how the operator and the agent work together, from the per-turn ledger the harness-dev hooks write — re-ask ratio and its causes, prompts to done, prompt quality, pasted blocks that should be a skill, and "You should know" notes. Use for "agent metrics", "how are my prompts", "re-ask ratio",

• help-command

  Use when writing or fixing a `<router>:help` slash command in a plugin under harness/plugins/ — a command whose only job is to print a static help table/guide. Trigger on "add a help command", "write a help command", "the help command prints junk / shows load summary instead of the table", "fix code

• prompt-dry-run

  Use when the user runs "/harness-dev:prompt-dry-run <prompt>" or asks to draft/preview/dry-run how a prompt would be handled — reflect back the agent's reading of the prompt (intent, planned actions, limits) BEFORE acting, so misalignment surfaces first. Trigger on "draft this", "what would you do w

• prune-text

  Rewrite a skill, an AGENTS.md/CLAUDE.md, or a doc corpus into the agent-writing house style — cut duplication across files, cut dead weight inside each file, then reshape into steps plus disclosed reference. Use when the user asks to prune, dedup, compress, or restyle docs or skills, or says files r

• repo-settings

  Set up a repository's base documentation and Claude configuration — README.md, CLAUDE.md, a per-repo deploy skill, a per-repo adr skill, and .claude/settings.json. Use for "set up this repo", "add repo settings", "bootstrap README and CLAUDE.md", "add a deploy skill to this repo", "set up ADRs here"

• rule-conflicts

  Find contradicting and duplicate rules across the instruction corpus — global CLAUDE.md, loose and plugin skills, ~/.notes/rules. Triggers on "find contradictions in the rules", "which rules conflict", "check the corpus for duplicates", and after a session broke a rule that another rule seemed to al

• session-cost-audit

  Audit why a Claude Code session was slow or expensive, and rank fixes onto existing harness levers. Triggers on "make sessions faster", "why did that session burn so many tokens", "where did the context go", "analyse this transcript", "too many turns per prompt", or any ask to profile a .jsonl under

• text-style

  Shape agent-facing markdown into the house style — steps plus disclosed reference, a checkable completion criterion per step, leading words instead of restatements, positive targets instead of bans, and a description that fires as a context pointer. Load it when writing or rewriting a SKILL.md, an a

Subagents (2)

• prune-blind-reader

  Over-cut detector for the `prune-text` skill — runs a pruned file as its only instruction, cold, and reports every point where it had to guess. Sees the pruned file and nothing else: no original, no cut list, no diff. Its guesses are the caller's over-cuts. Read-only on the corpus. Spawned once per

• prune-mapper

  Bounded read-and-report worker for the `prune-text` skill — reads a batch of files and emits one artifact: a topic map (phase 1) or an idea/property table (phase 2). Writes the artifact to a path and returns that path alone. Read-only on the corpus: it names what is there and never judges what to cu

Hooks (2)

• UserPromptSubmit

  python3 ${CLAUDE_PLUGIN_ROOT}/bin/turn-ledger.py

• Stop

  python3 ${CLAUDE_PLUGIN_ROOT}/bin/turn-ledger.py