CalvinBackup/gstack
1
0
Fork 0
mirror of https://github.com/garrytan/gstack.git synced 2026-06-29 13:10:05 +02:00
Code Issues Packages Projects Releases Wiki Activity

Merge origin/main into /spec branch — retag v1.45.0.0 → v1.47.0.0

Browse Source 
main moved to v1.46.0.0 (gstack v2 foundation, eval-first floor across
51 skills) while this branch was at v1.45.0.0. v1.46 also reserved
v1.45.0.0 for the design daemon feature. Retag this branch's release
v1.45.0.0 → v1.47.0.0 so it lands cleanly on top of main.

Conflict resolutions:
- VERSION: 1.47.0.0 (MINOR continues on top of main's 1.46.0.0; this
  branch is also a MINOR per scale-aware rules — new skill capability).
- CHANGELOG: rewrite this branch's release header v1.45.0.0 → v1.47.0.0.
  Keep both main entries above main's older history.

Adapts to main's eval-first floor (v1.46.0.0 test/skill-coverage-matrix.ts
+ test/skill-coverage-floor.test.ts):
- Register /spec in SKILL_COVERAGE with 3 gate entries + 2 periodic.
- Skill catalog grows 51 → 52. Floor 6/6 structural checks pass.
- Catalog tokens: 4045 → 4116 (+71 for /spec, within v1.46's ≤7000 budget).
- Trim spec frontmatter description to single-paragraph block form to
  respect v1.46's catalog-trim intent (was 14 lines / ~900 chars,
  now 5 lines / ~350 chars; routing prose stays in body sections).
- 363/363 gate-tier tests pass across skill-coverage-floor (309) +
  skill-coverage-matrix (10) + skill-size-budget (3) + parity-suite (4) +
  spec-template-invariants (35) + spec-template-sync (2).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
Garry Tan
2026-05-26 18:54:21 -07:00
parent 33fdb2c9b2 22f8c7f4e1
commit 3d77d1edd6
132 changed files with 10945 additions and 4270 deletions
Download Patch File Download Diff File Expand all files Collapse all files
benchmark-models/SKILL.md
+13 -8
View File
@@ -2,14 +2,7 @@
name: benchmark-models
preamble-tier: 1
version: 1.0.0
description: |
Cross-model benchmark for gstack skills. Runs the same prompt through Claude,
GPT (via Codex CLI), and Gemini side-by-side — compares latency, tokens, cost,
and optionally quality via LLM judge. Answers "which model is actually best
for this skill?" with data instead of vibes. Separate from /benchmark, which
measures web page performance. Use when: "benchmark models", "compare models",
"which model is best for X", "cross-model comparison", "model shootout". (gstack)
Voice triggers (speech-to-text aliases): "compare models", "model shootout", "which model is best".
description: Cross-model benchmark for gstack skills. (gstack)
triggers:
- cross model benchmark
- compare claude gpt gemini
@@ -23,6 +16,18 @@ allowed-tools:
<!-- AUTO-GENERATED from SKILL.md.tmpl — do not edit directly -->
<!-- Regenerate: bun run gen:skill-docs -->
## When to invoke this skill
Runs the same prompt through Claude,
GPT (via Codex CLI), and Gemini side-by-side — compares latency, tokens, cost,
and optionally quality via LLM judge. Answers "which model is actually best
for this skill?" with data instead of vibes. Separate from /benchmark, which
measures web page performance. Use when: "benchmark models", "compare models",
"which model is best for X", "cross-model comparison", "model shootout".
Voice triggers (speech-to-text aliases): "compare models", "model shootout", "which model is best".
## Preamble (run first)
```bash