Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/fortunto2/rust-code/autoresearchnpx skills add fortunto2/rust-code --skill autoresearchgit clone --depth 1 https://github.com/fortunto2/rust-codeWhat 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 | $0.00073 | $0.00912 |
| Opus 5 | $0.00036 | $0.00456 |
| Sonnet 5 | $0.00015 | $0.00182 |
| Haiku 4.5 | $0.00007 | $0.00091 |
Grade A, and why
autoresearch 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 2d 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 — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/autoresearch — Self-Improving Prompt Optimization
Karpathy autoresearch pattern: Generate → Evaluate → Score → Keep/Discard → Mutate → Repeat
Architecture
Core engine: crates/sgr-agent/src/autoresearch.rs (Rust module, feature genai).
Dashboard: skills/autoresearch/dashboard.py (Python, standalone viewer).
Test cases: skills/autoresearch/test_cases/*.json (embedded at compile time via include_str!).
Targets
| Target | What it optimizes | Eval metric |
|---|---|---|
tool-selection |
Tool descriptions in agent | % correct tool chosen for 66 test tasks |
system-prompt |
Agent system prompt (SOUL.md) | Decision quality: right tool + coherent reasoning (4 criteria) |
skill |
Any SKILL.md file | Binary criteria pass rate (4 criteria) |
decision-parser |
Structured output schema | Parse success + field quality (5 criteria) |
Usage (Rust API)
use sgr_agent::autoresearch::{AutoResearch, Config, Target};
let config = Config {
target: Target::ToolSelection,
batch_size: 10,
cycle_secs: 120,
gen_model: "gemini-2.5-flash".into(),
eval_model: "claude-sonnet-4-6".into(),
data_dir: "autoresearch_data/tool-selection".into(),
};
let ar = AutoResearch::new(config);
ar.run(20).await?; // 20 cycles
Quick Start (CLI — when wired into rc-cli)
# Tool selection optimization (most impactful)
cargo run -- autoresearch tool-selection --cycles 20
# System prompt optimization
cargo run -- autoresearch system-prompt --cycles 10
# Optimize a specific skill
cargo run -- autoresearch skill --name delegate --cycles 15
# Optimize decision parsing
cargo run -- autoresearch decision-parser --cycles 10
# Dashboard (Python, standalone)
python3 skills/autoresearch/dashboard.py --target tool-selection --port 8501
How It Works
Each cycle:
- Generate N outputs with current prompt (via Gemini — same model agent uses)
- Evaluate each output against binary criteria (via Claude Sonnet — different model for objectivity)
- Score = sum of passed criteria across all outputs
- Keep if score > best_score, discard otherwise
- Mutate the winning prompt to try improvements
- Log to JSONL for dashboard tracking
- Wait for next cycle (default 2 min)
What ships with it
6 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 2d ago First seen · 93 lines · 73 tokens per session scan A b8b6a481aa53
autoresearch is a skill published in the GitHub repository fortunto2/rust-code (41 stars, last pushed 10d ago), licensed MIT. It adds 73 tokens to every session and 912 once invoked, about $0.0004 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 2026-08-30.
Other skills, from other repositories
test-repo
Use this skill to test strategy changes against a fresh test repository. Invoke when the user asks to "test against a test repo", "validate the changes", or wants to verify session hooks, commits, and checkpoint creation work correctly.
development
开发语言能力索引。Python、Go、Rust、TypeScript、Java、C++、Shell。当用户提到编程、开发、代码、语言时路由到此。.
model-update
新增或更新 CCX 项目中的模型注册技能。适用于 Claude 与非 Claude 模型，覆盖能力表、路由白名单、基准映射、代码生成与验证。.
conductor-implement
Executes the tasks defined in the specified track's plan. Use this to start or continue working on a feature, bug fix, or chore.
skills-audit
Audit .agents/skills SKILL.md files. Use for recurring checks of duplicate, overlapping, stale, inconsistent, or broken skills and merge/delete candidates.
bionic-decision-engine
Unified mathematical arbitrator for all resource allocation decisions — money, time, energy, relationships. Absorbs 46 decision protocols + 24 strategy protocols into one dense engine.