skills/seed/SKILL.md
Vendor the system-wide Harness Kit catalog into the current repo when a project needs checked-in local copies instead of relying on global bootstrap symlinks. Copies first-party skills and agents into a repo-local shared skill layer, then bridges harness-specific entrypoints back to that shared copy. Use when: "seed this repo", "vendor harness kit here", "initialize the agent here", "set me up offline". Trigger: /seed.
npx skillsauth add phrazzld/spellbook seedInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Vendor Harness Kit into this repo. Most repos should rely on system-wide
bootstrap instead; use /seed only when the repo needs checked-in local
copies for offline work, review, or repository-specific persistence.
Find the Harness Kit root: readlink -f this SKILL.md, walk up until you
see skills/ + agents/ + harnesses/.
Resolve the repo-local shared skill root. Use .agents/skills/.
If a repo already has legacy .agent/skills/, stop and ask whether to
migrate it. The shared root is the canonical storage for vendored
Harness Kit skills; .claude/skills/, .codex/skills/, and .pi/skills/
are bridges.
Copy every skill in <harness-kit-root>/skills/ into the shared skill root,
preserving each skill's full directory (references/, scripts/,
evals/, everything). Skip seed itself; it lives globally.
Ensure .claude/skills/<name>, .codex/skills/<name>, and
.pi/skills/<name> point at each shared skill when those harness dirs
exist. Do not duplicate skill contents into harness-specific dirs.
Copy every agent in <harness-kit-root>/agents/ into the repo's existing
agent directory. In most repos today that is .claude/agents/.
Do not invent a second copy of agents unless the repo already has
a documented shared-agent convention.
Copy <harness-kit-root>/harnesses/shared/AGENTS.md to ./AGENTS.md
only if one doesn't already exist.
Print what you installed.
~/.claude / ~/.codex / ~/.pi.
Writes only to the current repo..claude/skills/ is only a bridge..claude/, or
AGENTS.md — ask first./seed exists only for repo-local
vendoring.Typical time: seconds. Typical cost: zero LLM tokens beyond the skill body — just file copies.
development
Lightweight evidence-backed retro and catch-up reports for a current repo, branch, PR, backlog slice, or recent agent session. Use when the user asks for a debrief, catch me up, what changed, why it matters, product implications, end-user implications, developer experience implications, current app state, backlog state, workspace state, alternatives considered, or context rebuild after losing the thread. Trigger: /debrief.
testing
Capture agent-session work records as local JSONL audit evidence. Links a backlog/spec, branch, commits, review verdicts, QA/demo evidence, transcript refs, and shipped ref without storing raw private transcripts. Use when: "trace this work", "write work record", "agent session trace", "journal this delivery", "link transcript evidence". Trigger: /trace, /journal.
data-ai
Turn proven agent-session patterns into first-party Harness Kit skills. Use when: "skillify this conversation", "make this into a skill", "generate a skill from current transcript", "extract reusable workflow". Trigger: /skillify.
testing
Run one targeted, read-only architecture or quality critique through a named lens from the shared rubric. Use when: "critique this module", "run an Ousterhout pass", "lens critique", "architecture critique". Trigger: /critique.