Skills Manager CLI

Before doing anything

1. Resolve the CLI first, then use the path it prints. Run this once (POSIX shell):

   D="$HOME/.skills-manager/bin"
   B="$D/skills-manager-cli"; [ -e "$B" ] || B="$B.exe"   # .exe on Windows
   if [ -s "$D/.version" ] && [ -x "$B" ]; then
     echo "$B"
   elif [ -s "$D/.version" ] || [ -e "$B" ]; then
     echo BRIDGE_BROKEN
   else
     P="$(command -v skills-manager-cli 2>/dev/null || true)"
     [ -x "$P" ] && echo "$P"
   fi

   Substitute the printed path into every command below, wherever the examples write $SM. Do not carry $SM as a shell variable: each command you run is a new shell, so an assignment made here is gone by the next one.

   The three outcomes:

   • A path under ~/.skills-manager/bin — the desktop app published this copy, and the .version stamp appears only after it has been verified, so it always matches the app the user is running. Use it.
   • BRIDGE_BROKEN — something the app left behind is here but does not add up: an unstamped binary, or a stamp with no binary beside it. Either is what a copy that failed half-way leaves. Stop. Do not go looking for another CLI: that binary may predate a safety fix, and the machine has a desktop app whose version nothing here can match. Ask the user to open the Skills Manager app once, which republishes it.
   • A path from PATH — nothing was ever published here, so there is no stale copy to worry about: this is a CLI-only machine (a server install, a standalone download, a hand-built binary). Use it, but note it can be older than a desktop app if one is also installed.

   If nothing is printed at all, this skill doesn't apply — fall back to find-skills, or tell the user to install Skills Manager.

2. Always pass --json when you parse output yourself. Pretty-printed output is for the user; JSON is for you. Errors include ok=false, a stable code, and message on stderr with a non-zero exit code.

"$SM" --json skills list

When a deployment is refused

A deploy that would overwrite something that is not ours is refused outright — nothing at those paths is deleted, and nothing else in the batch is applied. That failure is machine-readable, so report the actual paths rather than the sentence:

{"ok": false, "code": "TARGET_CONFLICT", "kind": "target_conflict",
 "message": "Refusing to deploy: 1 of 2 target(s) …",
 "details": {"conflicts": [{"path": "/Users/me/.claude/skills/db",
                            "reason": "is not a managed deployment"}]}}

Tell the user which path is in the way, that its contents are untouched, and offer the two ways out: adopt it into the library (skills adopt), or move it aside and retry. Never delete it for them.

Mental model

There's one central library at ~/.skills-manager/skills/ that all agents share. Each skill has source metadata, preset membership, tags, and zero or more real deployments in agent directories. A preset is a reusable group; several presets may be deployed at the same time.

Keep these three states separate:

• Library: install/remove controls whether Skills Manager owns the skill.
• Preset membership: presets add-skill/remove-skill organizes the library only.
• Deployment: skills deploy/undeploy and presets deploy/undeploy control what an agent can actually see.

Internally, presets are still stored as scenarios for backward-compatible Git Backup. The CLI and UI call them presets.

Install

# From skills.sh marketplace
"$SM" skills install vercel-labs/agent-skills@react-best-practices

# Any git URL (use /tree/branch/subpath form when the skill lives in a sub-directory)
"$SM" skills install https://github.com/anthropics/skills.git
"$SM" skills install https://github.com/foo/bar/tree/main/skills/baz

# Local folder
"$SM" skills install ./my-skill

# Force a source type when the ref is ambiguous
"$SM" skills install foo/bar --skillssh
"$SM" skills install ./looks-like/owner-repo --local

Default is library-only — the skill enters the DB but doesn't appear in any agent yet. Prefer an explicit follow-up deployment so scope is unambiguous:

"$SM" skills deploy <skill> --agent claude_code --agent codex

--sync and --sync-preset remain legacy shortcuts for the exclusive active-preset workflow.

Ref resolution is deterministic, no path-existence guessing:

1. Starts with ./, ../, /, or ~/ → local path
2. Contains ://, ends in .git, or starts with git@ → git URL
3. Matches owner/repo, owner/repo/skill, or owner/repo@skill → skillssh
4. Otherwise → error; pass --local / --git / --skillssh to disambiguate

Always verify after install with skills list or skills show <name> so you can confirm the skill landed and report the preset / sync state back to the user.

Search

"$SM" --json skills search "react performance" --limit 5

Each result has install_ref (paste straight into skills install), installs (popularity proxy), and skills_sh_url. Show the top 1–3 with install counts before installing — anything with 10K+ installs is battle-tested; anything under 100 needs a careful look at the source repo.

Update / Check

# Re-fetch one skill (git/skillssh re-clones, local/import re-imports source dir)
"$SM" skills update <skill-name-or-id>

# Re-fetch all eligible skills
"$SM" skills update --all

# Just probe remote revisions, don't touch files
"$SM" skills check --all

check is the dry-run partner of update. Local-only skills (no git source) are reported as skipped: true.

An update replaces the skill's directory wholesale, so anything written inside it that the new version does not have would be destroyed. When the CLI detects that, it applies nothing and reports the paths instead:

{ "name": "ppt-master", "refreshed": false,
  "held_back_removals": ["library: templates/mine.pptx"] }

The field is omitted entirely when nothing is held back, so test for its presence rather than for an empty array. refreshed: false with held_back_removals is not a failure and not something to retry — the skill is untouched and still on its old version. Show the user the listed paths and ask. There is no CLI flag to override this; only the desktop app can confirm and proceed, because only a person can say those files are expendable. The paths are prefixed with where they live (library, or an agent key for a deployed copy).

Note what this does not cover: a file the user edited that the new version also ships is reported as surviving, because its path survives — the update overwrites their edits silently. Warn anyone keeping local modifications inside a skill folder.

Remove

# Always preview first when removing more than one
"$SM" skills remove <skill> --dry-run

# --yes is required for the actual delete; --json mode does NOT auto-confirm
"$SM" skills remove <skill> --yes

Remove deletes the central-library copy, all synced targets across agents, and the DB row. It's not reversible without re-installing.

Deploy / Undeploy

"$SM" skills deploy <skill> --agent claude_code
"$SM" skills undeploy <skill> --agent codex
"$SM" skills deploy <skill-a> <skill-b> --agent codex --dry-run
"$SM" skills deploy <skill> --agent claude_code --agent codex
"$SM" --json skills status <skill>

These commands change real managed deployments without deleting the central-library copy or changing preset membership. skills enable/disable are deprecated compatibility commands and do not change deployment; never use them.

skills deploy and skills undeploy always require at least one explicit --agent, whether the command names one skill or several. skills status also reports target rows left by a custom agent that is no longer registered, so stale deployments stay visible and can be cleaned with an explicit undeploy while the row exists.

Legacy exclusive sync

# Sync current active preset to all enabled agents
"$SM" skills sync

# Preview the target list — safe, no writes
"$SM" skills sync --dry-run

# Switch the one legacy active preset, then sync
"$SM" skills sync --preset "Web Dev"

# Only sync to a single agent (useful when one agent's directory got out of sync)
"$SM" skills sync --tool claude_code

Adopt skills installed elsewhere

When skills already live in an agent's directory (e.g. installed via npx skills add or manual git clone) but aren't in the central library, pull them in:

# Dry-run scan first — lists candidates without writing
"$SM" skills adopt ~/.claude/skills --dry-run

# Adopt everything found — each becomes source_type=local (can't auto-update from git)
"$SM" skills adopt ~/.claude/skills

# Adopt a single skill and pin it to a git source so `update` works later
"$SM" skills adopt ~/.claude/skills/react-best-practices \
  --git-url https://github.com/vercel-labs/agent-skills/tree/main/react-best-practices

# Or pass --git-subpath explicitly when the URL is just the repo root
"$SM" skills adopt ~/.claude/skills/react-best-practices \
  --git-url https://github.com/vercel-labs/agent-skills \
  --git-subpath react-best-practices

# Skill lives at the repo root? Pass an empty subpath
"$SM" skills adopt ~/.claude/skills/my-skill \
  --git-url https://github.com/me/my-skill --git-subpath ""

adopt auto-excludes anything already in the DB or already a sync target, so it's safe to re-run. --git-url requires either a URL with a subpath (/tree/branch/path) or an explicit --git-subpath — without that, future update would re-clone the wrong directory, so the CLI refuses to guess.

--git-url only applies at the moment of adoption, while the directory is still unmanaged. Once a skill is in the library, use set-source below.