Install
$ agentstack add skill-bensheridanedwards-architectplaybook-install-architect-playbook-locally ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
/install-architect-playbook-locally
Copy every skill folder from this architect-playbook repository into /.claude/skills/. After running this, every architect-playbook slash command works inside the current project, and only inside the current project. Other projects on the machine are untouched.
Usage
/install-architect-playbook-locally # install or update every skill into .claude/skills/
/install-architect-playbook-locally --dry-run # print the plan without copying anything
/install-architect-playbook-locally --force # overwrite destinations even if they appear newer
/install-architect-playbook-locally --include=name # install only one named skill (repeatable)
/install-architect-playbook-locally --exclude=name # skip a named skill (repeatable)
What this skill does
- Resolves the playbook root. Locate the architect-playbook repository the user wants to install from. The playbook root is the directory that contains this skill's parent folder (
install-architect-playbook-locally/SKILL.md). If the skill is being run from inside~/.claude/skills/install-architect-playbook-locally/SKILL.md(already globally installed), ask the user for the path to the playbook clone. - Resolves the destination.
/.claude/skills/. Create it if it does not exist. - Enumerates source skills. Every direct sub-folder of the playbook root that contains a
SKILL.md, exceptinstall-architect-playbook-locallyandinstall-architect-playbook-globallythemselves (those should not be installed into target projects — they belong to the playbook). - Copies each skill folder with
cp -R / /. The destination becomes a complete, self-contained copy. Subsequent edits to the playbook do not propagate; re-run this skill to refresh. - Reports the result with one line per skill:
installed,updated,skipped (stub), orskipped (newer at destination).
Implementation steps
Step 1 — Resolve the playbook root
# This skill file lives at /install-architect-playbook-locally/SKILL.md when running
# from the source repository. Detect that case first.
SKILL_DIR="$(cd "$(dirname "${BASH_SOURCE[0]:-$0}")" && pwd 2>/dev/null || pwd)"
PLAYBOOK_ROOT="$(dirname "$SKILL_DIR")"
if [ ! -f "$PLAYBOOK_ROOT/install-architect-playbook-locally/SKILL.md" ]; then
echo "Could not auto-detect the architect-playbook root."
echo "Re-run from inside a clone of architect-playbook, or pass --from=."
exit 1
fi
When invoked as a slash command (not as a shell script), use the Read tool to find the playbook clone the user pointed at, or ask the user for the path.
Step 2 — Resolve destination
DEST="$(pwd)/.claude/skills"
mkdir -p "$DEST"
Step 3 — Enumerate skills to install
For each direct sub-folder of $PLAYBOOK_ROOT that contains a SKILL.md, build a list. Exclude:
install-architect-playbook-locallyinstall-architect-playbook-globally- Any folder whose name begins with
.
Apply --include and --exclude filters from the command line.
Step 4 — Detect stubs
A SKILL.md is considered a stub if its body contains the literal string **Status:** stub. Stubs are still copied by default (so they are visible as slash commands and the user knows what is on the way), but the report flags them with skipped (stub) if --no-stubs is passed.
Step 5 — Copy each skill
for skill in "${SKILLS[@]}"; do
src="$PLAYBOOK_ROOT/$skill"
dst="$DEST/$skill"
if [ -d "$dst" ] && [ -z "$FORCE" ]; then
# Compare modification time; skip if destination is newer.
if [ "$(stat -f %m "$dst/SKILL.md" 2>/dev/null || stat -c %Y "$dst/SKILL.md")" \
-gt "$(stat -f %m "$src/SKILL.md" 2>/dev/null || stat -c %Y "$src/SKILL.md")" ]; then
echo "skipped (newer at destination): $skill"
continue
fi
fi
rm -rf "$dst"
cp -R "$src" "$dst"
echo "installed: $skill"
done
Step 6 — Print the summary
installed: pre-audit-setup
installed: system-self-improve
installed (stub): security-audit
installed (stub): performance-audit
...
15 skills installed into ./.claude/skills/
Open a new Claude Code chat in this directory to pick up the new slash commands.
Idempotency rules
- Re-running with no flags is safe. Files only change if the source is newer (or
--forceis set). - This skill never deletes a destination skill that is not in the source — manual cleanup only.
- This skill never touches
~/.claude/skills/. For machine-wide install, use/install-architect-playbook-globally.
Safety
- Refuses to write outside
/.claude/skills/. - Refuses to follow symlinks out of the playbook root when copying.
--dry-runis honored — when set, the skill only prints the plan.
Recommended commit message
If the user wants to track the installed skills in version control:
chore: install architect-playbook skills into .claude/skills/
What this skill explicitly does NOT do
- Install graphify (run
/pre-audit-setupfor that). - Modify settings.json (run
/pre-audit-setup). - Run any audit.
- Affect any project other than the current working directory.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: BenSheridanEdwards
- Source: BenSheridanEdwards/ArchitectPlaybook
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.