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.
git clone --depth 1 https://github.com/smrafiz/claude-superchargerWrote 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/commands/smrafiz/claude-supercharger/sc)<a href="https://agentmods.dev/commands/smrafiz/claude-supercharger/sc"><img src="https://agentmods.dev/badge/commands/smrafiz/claude-supercharger/sc/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/commands/smrafiz/claude-supercharger/sc"><img src="https://agentmods.dev/badge/commands/smrafiz/claude-supercharger/sc.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 | $0.00000 | $0.00886 |
| Opus 5 | $0.00000 | $0.00443 |
| Sonnet 5 | $0.00000 | $0.00177 |
| Haiku 4.5 | $0.00000 | $0.00089 |
Grade B, and why
sc scanned grade B with 1 finding 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 9d 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
- `off` also moves Supercharger's **own** MCP servers aside — the `#supercharger`-tagged entries in both `~/.claude.json` and `~/.claude/settings.json` — so they stop loading and stop costing context; `on` restores each What it actually says
Activate or deactivate Claude Supercharger. Arguments: $ARGUMENTS (off | on | status)
Deactivate to get plain default Claude Code behavior on demand; reactivate when you want the guards, memory, economy, and statusline back. Nothing is uninstalled — off just switches everything off and keeps the files dormant on disk so on can restore them.
Run it (pass the argument straight through; default to status if none given):
bash ~/.claude/supercharger/tools/sc-toggle.sh $ARGUMENTS
Then report the tool's output verbatim.
/sc off— switch to default Claude behavior. Sets a global kill-switch so every hook exits immediately (no enforcement, no context injection, no statusline), removes the Supercharger block from~/.claude/CLAUDE.md, and moves Supercharger's own~/.claude/rules/*.mdaside. Security note: while off, ALL guards are inactive — destructive-command blocking, path-guard, credential/env-file guards, git-safety. State this clearly to the user./sc on— re-enable Supercharger: removes the flag, restores the CLAUDE.md block and the rules files. Hooks are active on the next tool call; the prompt rules re-enter context next session./sc status— report whether Supercharger is ACTIVE or DISABLED.
Notes to surface to the user:
- Hooks deactivate immediately (next tool call). The CLAUDE.md prompt layer changes take effect next session (it's loaded once at session start).
offalso unregisters Supercharger's hooks fromsettings.json, so they no longer even spawn. The kill-switch flag is read inside the hook body, so with the flag alone a disabled hook still forked, sourced its lib and exited — ~18 process spawns per Bash call, costing bash startup on every one. Only#supercharger-tagged entries are removed; hooks you registered yourself stay.onputs them back and verifies the count against the file it just wrote — if any are missing it stops, keeps the saved copy, and leaves Supercharger off rather than reporting a success that would leave you unguarded. A plugin install registers through the plugin's ownhooks.json, which this cannot edit, so the flag still does that job.offwrites a timestamped backup under~/.claude/backups/before changing anything.offalso moves Supercharger's own~/.claude/agents/*.mdaside — they never run once the hooks are gone, but their descriptions are listed to the model every session (~2970 tokens). Identified against the reference copy in~/.claude/supercharger/agents/, so an agent you wrote is never touched; an install predating that copy moves nothing rather than guessing. The 48 slash commands are deliberately kept:/scis one of them, so removing them could leave you with no way to type/sc on, and the whole set costs ~650 tokens.offalso moves Supercharger's own~/.claude/rules/*.mdaside —supercharger.md,guardrails.md,economy.md,anti-patterns.ymland your active role file. Claude Code auto-loads that directory with no import, so stripping only the CLAUDE.md block used to leave ~9KB of Supercharger instructions entering every session while it claimed to be off. A rules file you wrote yourself is never touched; role files are identified against the installed sources in~/.claude/supercharger/roles/.onputs them back, and refuses to overwrite anything that took the name while off.offalso moves Supercharger's own MCP servers aside — the#supercharger-tagged entries in both~/.claude.jsonand~/.claude/settings.json— so they stop loading and stop costing context;onrestores each one to the file it came from. Your own MCP servers are never touched. MCP loads at session start, so this takes effect next session.
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.
- 9d ago First seen · 24 lines · 0 tokens per session scan B eab7cf73929e
sc is a command published in the GitHub repository smrafiz/claude-supercharger (11 stars, last pushed today), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 886 tokens. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.