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 agentmods add commands/tzachbon/smart-ralph/switchgit clone --depth 1 https://github.com/tzachbon/smart-ralphWhat 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 | $0.00003 | $0.00498 |
| Opus 5 | $0.00002 | $0.00249 |
| Sonnet 5 | $0.00001 | $0.00100 |
| Haiku 4.5 | $0.00000 | $0.00050 |
Grade A, and why
switch 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 2d 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.
What it actually says
Switch Active Feature
You are switching the active feature.
Parse Arguments
From $ARGUMENTS:
- name: The feature name to switch to (required)
Validate
- If no name provided, list available features and ask user to choose
- Check if
.specify/specs/$name/exists (supports both###-nameand plainnamematching) - If not, error: "Feature '$name' not found. Available features: "
List Available (if no argument)
If $ARGUMENTS is empty:
- List all directories in
.specify/specs/ - Read current active feature from
.specify/.current-feature - Show list with current marked
Available features:
- 001-user-auth [ACTIVE]
- 002-payment-flow
- 003-dashboard
Run: /speckit:switch <feature-name>
Execute Switch
-
Find matching feature directory (exact match or partial match on name portion):
# Check for exact match first if [ -d ".specify/specs/$name" ]; then FEATURE="$name" else # Try matching by name suffix (e.g., "user-auth" matches "001-user-auth") FEATURE=$(ls -1 .specify/specs/ 2>/dev/null | grep -E "^[0-9]{3}-$name$" | head -1) fi -
Update
.specify/.current-feature:echo "$FEATURE" > .specify/.current-feature -
Read the feature's state:
.speckit-state.jsonfor phase and progress.progress.mdfor context
Output
Switched to feature: $FEATURE
Current phase: <phase>
Progress: <taskIndex>/<totalTasks> tasks
Files present:
- [x/blank] spec.md
- [x/blank] plan.md
- [x/blank] tasks.md
- [x/blank] .progress.md
Next: Run /speckit:<appropriate-phase> to continue
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.
- 2d ago First seen · 77 lines · 3 tokens per session scan A 0074a5fc74c9
switch is a command published in the GitHub repository tzachbon/smart-ralph (527 stars, last pushed 5d ago), licensed MIT. It adds 3 tokens to every session and 498 once invoked, about $0.0000 per session on Opus 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 2026-08-30.
Other commands, from other repositories
loop
Iteratively fix issues until all resolved or max iterations reached.
project
Generate project documentation (product.md, structure.md, tech.md, codemaps/).
version
Display current guide and Claude Code versions.
release
This compatibility command keeps /oh-my-claudecode:release available without loading the full release skill description in every Claude Code session.
router-on
Route Claude Code through the Weave Router again (turn it back on).
auto-run
PitWay: Manage auto-run authorization for automatic task continuation.