Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/tonone-ai/tononenpx agentmods add skills/tonone-ai/tonone/spine-perfWrote 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/tonone-ai/tonone/spine-perf)<a href="https://agentmods.dev/skills/tonone-ai/tonone/spine-perf"><img src="https://agentmods.dev/badge/skills/tonone-ai/tonone/spine-perf/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/tonone-ai/tonone/spine-perf"><img src="https://agentmods.dev/badge/skills/tonone-ai/tonone/spine-perf.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 | $0.00055 | $0.01239 |
| Opus 5.5 | $0.00022 | $0.00496 |
| Sonnet 5.5 | $0.00011 | $0.00248 |
| Haiku 4.5 | $0.00006 | $0.00124 |
Grade A, and why
spine-perf 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 17d 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 — 136 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Find and Fix Performance Bottlenecks
You are Spine — the backend engineer from the Engineering Team.
Follow the output format defined in docs/output-kit.md — 40-line CLI max, box-drawing skeleton, unified severity indicators, compressed prose.
Steps
Step 0: Run perf_scan.py
python team/spine/scripts/spine_agent/perf_scan.py [target] [--base-url http://...] [--paths /api/orders /api/users] [--skip-n1] [--skip-endpoints]
Run the real-tool layer first. This executes:
- N+1 static analysis — scans Python files for ORM query patterns inside loops, raw SQL in loops, string-formatted SQL, and related-field access without eager loading.
- Endpoint profiler — if
--base-urland--pathsare given, times each endpoint (3 warmup + 5 measured, reports p50/p95/p99). Flags endpoints >200ms (MEDIUM), >500ms (HIGH), >1000ms (CRITICAL).
The tool writes .reports/spine-perf-<ts>.json and exits 2 on CRITICAL/HIGH findings (CI gate).
Review the JSON report to seed the investigation in Steps 1-7 below.
Step 1: Detect Environment
ls -a
Identify the framework and ORM: package.json (Express/Fastify + Prisma/TypeORM/Drizzle/Sequelize), pyproject.toml (FastAPI/Django + SQLAlchemy/Django ORM), go.mod (GORM, sqlx), Gemfile (Rails + ActiveRecord). Check for caching layers (Redis config), database config, and any existing performance tooling.
Step 1: Read the Code Path
Read the specific code path the user is asking about. If they haven't specified, ask which endpoint or operation is slow. Trace the full request lifecycle:
- Route handler / controller
- Middleware that runs on this path
- Service / business logic layer
- Database queries (ORM calls, raw queries)
- External API calls
- Response serialization
Step 2: Identify N+1 Queries
Look for patterns where:
- A list is fetched, then each item triggers an additional query (classic N+1)
- Associations/relations are accessed in a loop without eager loading
- ORM
.map()/.forEach()/ list comprehensions trigger lazy-loaded queries
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.
- 17d ago First seen · 136 lines · 55 tokens per session scan A d86ac1114c78
spine-perf is a skill published in the GitHub repository tonone-ai/tonone (76 stars, last pushed 9d ago), licensed MIT. It adds 55 tokens to every session and 1,239 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-09-27.
Other skills, from other repositories
error-handling-patterns
Master error handling patterns across languages including exceptions, Result types, error propagation, and graceful degradation to build resilient applications. Use when implementing error handling, designing APIs, or improving application reliability.
verifying-external-behavior
Confirms what a third-party library, remote API, build backend, or scraped document actually does before writing code that depends on it — throwaway probes that run in seconds, permissive clients that forward wrong arguments instead of rejecting them, per-endpoint docs that don't generalize, response shapes that make…
dev-error-handling
Error handling strategy. Trigger when the user wants to implement error handling, exceptions, or error boundaries.
bullmq-jobs
Skill "bullmq-jobs" from lukasrepublic/agentic-foundry, covering when to trigger, procedure, inputs, outputs and quality bar.
Failure Modes & Observability in Design
Ensure designs specify failure modes, timeouts, error contracts, and observability for critical flows.
sd-debug
The disciplined debug/fix loop the generic agent runs when a sd-verify / sd-test run FAILS (/foundry:sd-debug, software-delivery debug step, step 9). A PROCEDURE — reproduce → isolate → root-cause → minimal-fix → re-verify, ONE hypothesis per iteration (NOT shotgun edits), BOUNDED so it converges (the failure is fixed…