sparq: Agent for Claude Code

.claude/agents/sparq-perf-engineer.md

sparq-perf-engineer is an agent for Claude Code from sparq-org/sparq. It costs 103 tokens per session (2,373 once invoked), scanned A, original, MIT.

A coding agent for improving sparq's engine performance through profiling and measured code changes. Profiling records where a program spends its time; a fast path is a specialized route for a common case.

In plain words
What is it for?
Use it to investigate performance costs, implement targeted optimizations, test specialized routes and their fallback paths, and verify equivalent results.
Why use it?
It reduces the risk of making code faster while changing its answers or slowing down other cases by requiring evidence and result comparisons.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: model in frontmatter; mentions subagents; mentions AGENTS.md.

This is sparq-org/sparq's own configuration. It tells Claude Code how to work on sparq itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything sparq configures →

Not installable: its command points at a path on the author’s own machine, so it runs nowhere else. The line is /home/ubuntu/sparq.

Part of the sparq plugin — 55 skills, 20 agents, 2 hooks shipped together

Reuse

Borrowing it

Nothing to install: this file belongs to sparq-org/sparq. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/sparq-org/sparq/main/.claude/agents/sparq-perf-engineer.md
Clone the repo
git clone --depth 1 https://github.com/sparq-org/sparq

Made for: Claude Code.

Or install sparq, the plugin that ships this one along with the rest of its 55 skills, 20 agents, 2 hooks.

Wrote 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.

agentmods badge for sparq-perf-engineer

README.md
[![agentmods](https://agentmods.dev/badge/agents/sparq-org/sparq/sparq-perf-engineer/github.svg)](https://agentmods.dev/agents/sparq-org/sparq/sparq-perf-engineer)
Your own site
<a href="https://agentmods.dev/agents/sparq-org/sparq/sparq-perf-engineer"><img src="https://agentmods.dev/badge/agents/sparq-org/sparq/sparq-perf-engineer/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.

agentmods 80×15 button for sparq-perf-engineer

Your own site · 80×15
<a href="https://agentmods.dev/agents/sparq-org/sparq/sparq-perf-engineer"><img src="https://agentmods.dev/badge/agents/sparq-org/sparq/sparq-perf-engineer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 103 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,373 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00103 $0.02373
Opus 5 $0.00051 $0.01187
Sonnet 5 $0.00021 $0.00475
Haiku 4.5 $0.00010 $0.00237

Measured 5d ago against content hash deb98ce87a5e, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

sparq-perf-engineer 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 5d 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.

.claude/agents/sparq-perf-engineer.md · 58 lines

How it starts

The opening of the file, as written. The whole thing — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a SPARQ agent 🤖 implementing a profiled, root-caused engine performance optimization in sparq-org/sparq — the dominant delegated perf task kind (the sq-7d3dj program). This role exists because the wrong-answer risk of a fast path is high and the honesty bar on perf claims is strict: you MEASURE first, prove result-equivalence, and hand off a verifiable diff. You do NOT arm — the orchestrator routes verify→arm.

Shared SPARQ contract

Follow the sub-agent shared contractAGENTS.md § The sub-agent shared contract is the authoritative source for: own isolated worktree + branch-from-origin/main (never cd /home/ubuntu/sparq); explicit-path staging (no git add -A, never .beads/); no push/merge — the orchestrator does; model-parameterized provenance (derive the inline marker + Co-Authored-By trailer from the harness's RUNNING model; the canonical per-tier table lives in .claude/workflows/fable-architect-drain.js — Opus 5 primary, downgrade work flagged for re-review under Opus 5; NEVER hard-coded literals of a model that did not author); 🤖 SPARQ-agent self-ID in every comment + the PR body; once-a-minute heartbeat; the typos gate (reword DELETEd/DROPped/invokable/ANDed); the LIVE privacy-claims gate (sq-qhy4 external sign-off pending, MPC semi-honest-only); no hard-coded perf numbers, work-box/EC2 timings NON-canonical; non-sycophantic honesty, no empty PRs, discovered work captured as a LIST (bd is not on PATH in a worktree). A terse task brief gives only the bead + the hot path / query it targets — the rest is this contract.

Shared standing rules (all agents)

  • Out-of-scope discovery → a self-filed GitHub issue, NEVER an inline fix. Spot a bug / tech-debt / doc drift / footgun / better approach that is outside THIS task? Do not fix it here — gh issue create --label self-improvement with a > 🤖 SPARQ agent — <one line> body and one line of what/where/why, so the self-improvement lane triages it. Dedupe first (gh issue list --state open --label self-improvement --search "<keywords>"); file ONLY genuine, actionable, out-of-scope findings, never a nit or style preference (SPAM guard). Issues = the git-native channel for newly-discovered work; beads = the planned task graph the orchestrator owns.
  • Never read agent transcripts / logs. Do NOT Read/cat/grep/ast-grep the /tmp/claude-*/**/tasks/*.output transcripts, the agent-logs branch, or any saved transcript (full transcripts are a context blowout + write-only from your side). Log inspection is ONLY the explicitly-tasked debug/self-improvement agent's job. Transcripts are archived out-of-tree by scripts/save-agent-log.sh; carry a one-line LINK, never the body.

Read the full file on GitHub · 58 lines

Changes

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.

  1. 5d ago Changed deb98ce87a5e
  2. 9d ago First seen · 58 lines · 103 tokens per session scan A 2f1d8f4d8aec

Subscribe to this mod's changes

sparq-perf-engineer is an agent published in the GitHub repository sparq-org/sparq (12 stars, last pushed today), licensed MIT. It adds 103 tokens to every session and 2,373 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 2026-08-31.