Mibayy/token-savior

> One MCP server. One profile. **97.9% on tsbench at -80% tokens.**
> Structural code navigation, persistent memory, and Bash command rewriting for AI coding agents.
-brightgreen)
**mibayy.github.io/token-savior** -- project site + benchmark landing
Benchmark source + fixtures: not currently published (see *Reproducing the score* below)
v4.23 in 31 seconds, measured on 2026-10-07 · watch the full video (MP4, sound on)
---
Benchmark -- 96 real coding tasks (Claude Opus 4.7, May 2026)
| | Plain Claude Code | With Token Savior |
|---|---:|---:|
| **Score** | 141 / 180 (78.3%) | **188 / 192 (97.9%)** |
| **Active tokens / task** | 17 221 | **3 395** (-80%) |
| **Wall time / task** | 110.6 s | **18.9 s** (-83%) |
Reproduces with the optimized profile (single env var). The harness that
produced these numbers is described below; its repository is not public at the
moment, so take the figures as reported rather than as independently verifiable.
**A re-measurement was published here on 2026-08-09 and has been withdrawn on
2026-08-10.** It reported new-token savings from a small replacement harness.
Those numbers did not measure this server at all: across 143 benchmark
sessions, **exactly one** called a Token Savior tool. The client running the
harness had MCP deferred-tool loading enabled, so all 18 tools sat behind a
ToolSearch lookup instead of appearing in the model's manifest. The model
never saw them and fell back to Grep and Read — 66 greps, 30 reads, one
MCP call. What varied between the "profiles" was the size of the cached
prefix, not what the agent did.
The lesson is worth more than the numbers were: **a benchmark of an MCP server
must assert that its tools were actually called.** Ours did not, so it happily
compared two identical agents. That assertion now exists in the harness.
The headline figures above therefore stand as reported and unverified, as
stated in the previous paragraph.
**Re-measured on 2026-10-07, with the tools actually called.** Same 8 tasks
on a frozen copy of this repo (commit 9d15a1e, no CLAUDE.md), Claude
Haiku 4.5, plain Claude Code against the auto profile with alwaysLoad.
The harness checks that the Token Savior sessions really called the server.
| 3 passes, mean ± sd | Plain Claude Code | **v4.23.0 auto** |
|---|---:|---:|
| **Cost, 8 tasks** | $0.226 ± 0.011 | **$0.123 ± 0.020 (-45%, range -39% to -53%)** |
| **New tokens** | 92 609 ± 6 731 | **38 150 ± 7 618 (-59%)** |
| **Sessions that used TS** | — | **23 / 24** |
| **Wall time, 8 tasks** | 157 ± 26 s | 216 ± 9 s (+40%) |
All 48 sessions succeeded. v4.22.0, measured once, saved 14% and used the tools
in 5 sessions out of 8. Editing, measured on 2026-10-08 on the same repository (3 tasks in files of
1 900 to 3 700 lines, graded by running the code and the module's tests,
4 passes): every session of both arms succeeded, and cost is too noisy to call
a gain (from -54% to +16% per pass). Native Edit only needs a partial
Read, so there is little left to remove there.
Wall time is worse because each benchmark session starts a fresh server that
indexes the repo first; a long-lived session pays that once. What moved from
v4.22.0 to v4.23.0 is mostly wording: search_codebase now names the function
holding each hit, and the tool descriptions say plainly which native call they
replace.
---
Who starred this repo?
On June 30, 2026 GitHub restricted stargazer and watcher lists to repo admins and
collaborators, which broke every "who starred my repo" tool at once. I rebuilt one
that still works, precisely because it only reads repos you own or can push to:
**starscope** ranks the people who starred or
forked your repo by influence, and surfaces their social accounts when their
GitHub profile declares them.
Numbers on this very repo, computed with it: 1,147 people, 27% with a public
social account, and the most followed carries 18,922 followers. The named list is
visible to the repo owner and to nobody else — the public
page shows aggregates only.
---
What's new
Release notes live where they can't drift out of sync with the code:
- CHANGELOG.md for the full history
- Releases for the tagged builds
- PyPI for what pip will actually give you
---
Quick start
Add to your MCP config (e.g. Claude Code):
**"alwaysLoad": true** matters on Claude Code (2.1.28x and later): without
it, every tool of the server sits behind ToolSearch and the model reaches for