Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add derjochenmeyer/claude-code-craft --skill craft-project-dreamgit clone --depth 1 https://github.com/derjochenmeyer/claude-code-craftWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/derjochenmeyer/claude-code-craft/craft-project-dream)<a href="https://agentmods.dev/skills/derjochenmeyer/claude-code-craft/craft-project-dream"><img src="https://agentmods.dev/badge/skills/derjochenmeyer/claude-code-craft/craft-project-dream/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/derjochenmeyer/claude-code-craft/craft-project-dream"><img src="https://agentmods.dev/badge/skills/derjochenmeyer/claude-code-craft/craft-project-dream.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00046 | $0.03720 |
| Opus 5.5 | $0.00018 | $0.01488 |
| Sonnet 5.5 | $0.00009 | $0.00744 |
| Haiku 4.5 | $0.00005 | $0.00372 |
Grade A, and why
craft-project-dream scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 391 lines — stays where its author put it; the contents beside it link to each section on GitHub.
craft-project-dream
Like sleep, dream processes what accumulated. Forgets what does not matter. Strengthens what does. Leaves the project rested so the next session starts sharp, not sluggish.
Every dream is auditable. Before any change is applied, every file dream will touch is snapshotted to .craft/dreams/<timestamp>/before/. After the run, the new state lands in after/ alongside a human-readable diff.md and a quantified report.md. If a dream makes things worse, rollback is one cp away.
Run after active sprints, at the end of a productive week, or when Claude keeps re-asking questions that should already be settled.
Step 0: Assess readiness
CWD=$(pwd)
NOW=$(date +%s)
STAMP=$(date +%Y-%m-%d-%H-%M)
DREAM_DIR="$CWD/.craft/dreams/$STAMP"
# Portable mtime (macOS: stat -f, Linux/WSL: stat -c)
file_mtime() { stat -f "%m" "$1" 2>/dev/null || stat -c "%Y" "$1" 2>/dev/null || echo 0; }
echo "=== FILES ==="
for f in CLAUDE.md .craft/PROJECT_STATE.md .craft/BACKLOG.md .craft/HANDOVER.md .craft/DECISIONS.md .craft/DREAM_LOG.md; do
if [[ -f "$CWD/$f" ]]; then
lines=$(wc -l < "$CWD/$f" | tr -d ' ')
age_days=$(( (NOW - $(file_mtime "$CWD/$f")) / 86400 ))
echo " FOUND $f ${lines} lines ${age_days}d ago"
else
echo " MISSING $f"
fi
done
echo "=== MEMORY ==="
mem_count=$(find "$CWD/.craft/memory/" -maxdepth 1 -type f 2>/dev/null | wc -l | tr -d ' ')
echo " $mem_count files"
ls -t "$CWD/.craft/memory/" 2>/dev/null | head -5
echo "=== GIT ==="
commits_since_last_dream=$(git log --since="$(file_mtime "$CWD/.craft/DREAM_LOG.md" 2>/dev/null | xargs -I{} date -r {} '+%Y-%m-%d' 2>/dev/null || echo '90 days ago')" --oneline 2>/dev/null | wc -l | tr -d ' ')
echo " $commits_since_last_dream commits since last dream"
echo "=== SNAPSHOT TARGET ==="
echo " $DREAM_DIR"
If CLAUDE.md is missing and .craft/ does not exist: stop with "This project is not set up yet. Run /craft-project-setup first."
If the project looks dormant (0 commits since last dream): note it but continue — dormant projects still benefit from consolidation.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 391 lines · 46 tokens per session scan A 90e4f6a3025f
craft-project-dream is a skill published in the GitHub repository derjochenmeyer/claude-code-craft (2 stars, last pushed 6mo ago), licensed MIT. It adds 46 tokens to every session and 3,720 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
mempalace-task
Create, hand off, claim, execute, and close agent tasks through the MemPalace logstream. Use when the user wants to delegate work, prepare a ready-to-paste task for another agent, receive a MemPalace task id, or explicitly launch a supported headless agent in controlled mode.
knowledge-meeting-ingest
Ingest meeting transcripts from any recorder into Knowledge with attendee enrichment, timeline merge, supported decisions and commitments, and verification of substance and sequence. Summary-only input is not a verified meeting ingestion.
knowledge-automation
Create scheduled Knowledge ingestion automations with a verified worker, private-vault authority, and an end-to-end scheduled run. Use when asked to schedule or repair recurring ingestion from any source.
gsd-mempalace-capture
File a phase artifact into MemPalace; mirror decision facts into its temporal KG.
neat-freak
Knowledge and governance closeout: reconcile project docs, rule files (CLAUDE.md/AGENTS.md), authorized agent memory, and workspace residue with what the code and runtime actually do, so the next session or the next person starts from one current answer. Trigger when the user names "neat-freak", "洁癖", or "/neat" — and…
md-hive-sync
Munder Difflin hive sync — runs the start-of-task hive protocol steps: reads memory.md, checks inbox/ for new messages, and reminds you to record durable facts in memory.md and write coordination files before ending. Use when asked to "sync with the hive", "check my inbox", "hive status", or "hive sync". Proactively…