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 ww-w-ai/super-token-saver --skill s-compactgit clone --depth 1 https://github.com/ww-w-ai/super-token-saverWrote 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/ww-w-ai/super-token-saver/s-compact)<a href="https://agentmods.dev/skills/ww-w-ai/super-token-saver/s-compact"><img src="https://agentmods.dev/badge/skills/ww-w-ai/super-token-saver/s-compact/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/ww-w-ai/super-token-saver/s-compact"><img src="https://agentmods.dev/badge/skills/ww-w-ai/super-token-saver/s-compact.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high System Prompt Leakage · line 116 Skill contains instructions that could directly expose system prompts, internal rules, or hidden instructions to users or external parties.Fix: Remove any instructions that reveal, print, or output system prompts or internal rules. System instructions should never be exposed to end users.
- medium Rogue Agent · line 24 Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
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 | $0.00099 | $0.02610 |
| Opus 5 | $0.00049 | $0.01305 |
| Sonnet 5 | $0.00020 | $0.00522 |
| Haiku 4.5 | $0.00010 | $0.00261 |
Grade A, and why
s-compact 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 10d 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 — 122 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Generate a session handoff: a document that lets the NEXT session pick up with full context and zero re-derivation — in EITHER tool. The handoff is stored per project, not per tool, so a session ended in Claude Code is picked up by Codex and the reverse.
/s-continue restores the transcript (user↔assistant messages); this skill captures what the transcript does NOT hold and saves it where /s-continue auto-loads it.
ONE JOB: distill the non-dialogue knowledge. The transcript is already restorable — do not re-summarize it.
The two tools form a ladder of recovery:
/s-continuealone → fast, accurate context restore with no wasted/compacttokens (the transcript)./s-compact+/s-continueas a set → ALSO recovers the hidden results and process that never appeared in the dialogue (subagent findings, tool-output numbers, process lessons). That extra layer is THIS skill's entire reason to exist — so it must be captured in FULL, not gestured at.
A terse handoff is a FAILED handoff. The common failure mode is a handoff too short to carry the hidden layer — a few bullets that just restate what the transcript already has. This skill exists precisely because that is not enough. Favor completeness of the non-dialogue layer over brevity; the token cost of /s-compact is paid back many times over by the next session not re-running the same subagents and tool calls.
Help
ONLY if the user's argument literally contains "help" (e.g. /s-compact help). Otherwise skip to Step 1.
/s-compact — Write a session handoff for the next session (pairs with /s-continue)
/s-compact Generate the handoff, save it, print it, copy to clipboard
/s-compact help This help
The handoff is saved to ~/.claude/super-token-saver-data/<project>/handoff.md
On the next session, /s-continue restores the transcript AND auto-loads this file.
The path is shared by both tools, so the next session can be Claude Code or Codex.
Workflow: end a session → /s-compact · start the next → /s-continue
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 10d ago First seen · 122 lines · 99 tokens per session scan A 601fc337eca5
s-compact is a skill published in the GitHub repository ww-w-ai/super-token-saver (31 stars, last pushed 2d ago), licensed Apache-2.0. It adds 99 tokens to every session and 2,610 once invoked, about $0.0005 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 skills, from other repositories
cco-budget
Configure token budget limits, auto-compact settings, and view current budget status (model-aware — Claude 5 lineup, Opus 5 default fallback, full 1M context at standard price).
cco-clean
Clean up old tracking data and reset statistics.
agent-harness-optimizer
Use when agent harness optimization patterns for token efficiency, memory persistence, session management, and cross-harness parity. Use when optimizing agent performance, reducing token costs,.
git-workflow
Use when starting feature work that needs a branch, creating worktrees for isolation, making atomic commits during development, or completing a development branch via merge, PR, preserve, or discard. Also use when the user says "set up worktree", "create PR", "finish this branch", "start feature", or when you need an…
parallel-dispatch
Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies. Also use when multiple test files fail with different root causes, multiple subsystems are broken independently, or you want agents to work concurrently on separate problem domains. If tasks touch different…
pith
Adaptive output compression. Three levels: precise (no filler), lean (default, drop articles/fragments), ultra (maximum — abbreviate, structure, arrows). Activate: /pith | /pith precise | /pith lean | /pith ultra | /pith off.