Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add ppgranger/token-saver/plugin install 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/ppgranger/token-saver/token-saver-config)<a href="https://agentmods.dev/skills/ppgranger/token-saver/token-saver-config"><img src="https://agentmods.dev/badge/skills/ppgranger/token-saver/token-saver-config.svg" alt="Measured on agentmods" 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 | $0.00037 | $0.00991 |
| Opus 5 | $0.00018 | $0.00495 |
| Sonnet 5 | $0.00007 | $0.00198 |
| Haiku 4.5 | $0.00004 | $0.00099 |
Grade A, and why
token-saver-config scanned grade A 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 7d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
36 specialized processors, auto-discovered from `src/processors/`: git, gh (GitHub CLI), docker, kubectl, helm, terraform, pulumi, cdktf, ansible, nix, mise, npm/pip/cargo/go/maven/gradle/bun, package listing, python ins How it starts
The opening of the file, as written. The whole thing — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Token-Saver Configuration & Diagnostics
Check Status
Run token-saver stats to see lifetime and per-command compression statistics. Add --json for machine-readable output.
Check Why a Specific Command Was (or Wasn't) Compressed
There is no token-saver config subcommand — use explain and benchmark instead:
token-saver explain "git status" # routing decision: compressible? which processor? why excluded?
token-saver benchmark "npm install" --dry-run # measure compression without replacing real output
Configuration
Settings are read in this order (later wins): built-in defaults → ~/.token-saver/config.json → a project-level .token-saver.json found by walking up from cwd → TOKEN_SAVER_* environment variables. There is no CLI subcommand to write these files — edit ~/.token-saver/config.json directly, or set an environment variable.
The most commonly adjusted settings and their actual defaults:
enabled(bool, defaulttrue): master on/off switch.min_input_length(int, default1): minimum output length in characters before compression is attempted.min_compression_ratio(float, default0.0): a compression result below this fraction is discarded and the original output is returned instead.wrap_timeout(int, default300): max seconds a wrapped command may run before being killed.chars_per_token(int, default4): used only to estimate token counts for display (stats, benchmark) — not part of compression itself.disabled_processors(list, default[]): processor names (e.g."docker","kubectl") to turn off entirely.max_output_bytes(int, default10_000_000): hard cap on output length before compression even runs.
Example ~/.token-saver/config.json:
{
"min_compression_ratio": 0.1,
"disabled_processors": ["docker"]
}
Equivalent environment variables: TOKEN_SAVER_MIN_COMPRESSION_RATIO=0.1, TOKEN_SAVER_DISABLED_PROCESSORS=docker.
Security note: user_processors_dir, disabled_processors, and redaction_allowlist can only be set via the global ~/.token-saver/config.json or an environment variable — never via a project-level .token-saver.json (that file is auto-discovered from any directory you cd into, so it is not trusted with these).
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.
- 7d ago First seen · 61 lines · 37 tokens per session scan A 15a1f5cccb4d
token-saver-config is a skill published in the GitHub repository ppgranger/token-saver (142 stars, last pushed 19d ago), licensed Apache-2.0. It adds 37 tokens to every session and 991 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
michel-monitor-pull-request-github-actions
Diagnose a failed, stuck, or never-triggered CI run on a GitHub PR, apply a local fix if possible, push it, and document the result in a single running PR comment. Invoke whenever Michel's CI monitor loop triggers with anyfailure, stuck, or nottriggered — the bash loop already handles pending and allgreen silently, so…
resolving-merge-conflicts
Use when a git merge or rebase reports conflicts and the operation is in progress.
debugging
Debugging discipline: feedback loop FIRST, root cause before fix, blast radius after fix. Covers the 10-rung construction ladder, LSP-powered tracing, hypothesis quality criteria, and four-phase investigation. Loaded by bug-investigator.
codebase-hygiene
Two-mode skill: (1) find semantic duplicates — functions doing the same thing under different names, invisible to copy-paste detectors; (2) deepen shallow modules — thin wrappers and pass-through layers that spread complexity. Advisory and read-only; changes route through BUILD with full gates.
session-health-probe
Runtime daemon + live-session liveness probe (zero-LLM, deterministic): daemon up, sessions progressing (not wedged — RP41 double-signal), RSS under budget, no unrecovered failure events, deployed commit == expected. Snapshot now; runs as a 15-min job; red → Slack.
swarm-ci
Check SwarmAI GitHub Actions CI status: list recent runs, diagnose failures, and summarize health. Replaces ad-hoc gh run commands with structured output. TRIGGER: "CI status", "check CI", "is CI green", "CI failures". NOT FOR: pytest use cases.