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/d-oit/rust-2026-template/harnessnpx skills add d-oit/rust-2026-template --skill harnessgit clone --depth 1 https://github.com/d-oit/rust-2026-templateWhat 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 | $0.00068 | $0.00765 |
| Opus 5 | $0.00034 | $0.00382 |
| Sonnet 5 | $0.00014 | $0.00153 |
| Haiku 4.5 | $0.00007 | $0.00076 |
Grade A, and why
harness 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 — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill: harness
What Is the Harness
Agent = Model + Harness. The harness is the system of feedforward guides (what to do before coding) and feedback sensors (what catches violations after coding). It has two axes:
- Feedforward (guides): Context, constraints, conventions that prevent errors before they happen.
- Feedback (sensors): Automated checks that fire after code changes, providing structured error output.
The harness has two modes:
- Computational: Deterministic checks (fmt, clippy, deny) — always trust the output.
- Inferential: LLM-based guidance (skill docs, agent context) — direction, not commands.
Sensor Response Protocol
When a computational sensor fires:
- Read the full error message — it includes a fix hint.
- Classify the error: fmt / lint / test / arch / security.
- Apply the minimal fix — do not refactor unrelated code.
- Re-run the specific sensor —
cargo fmt,cargo clippy, etc. - Only commit when the sensor is green.
- Write a metrics event to
.agents/events/YYYY/MM/DD/per themetrics-reporterskill.
Sensor Quick Reference
| Sensor | Command | Config | Stage |
|---|---|---|---|
| fmt | cargo fmt --all -- --check |
.pre-commit-config.yaml |
pre-commit |
| clippy | cargo clippy --all-targets --all-features -- -D warnings |
.clippy.toml, .pre-commit-config.yaml |
pre-commit + CI |
| deny | cargo deny check |
deny.toml |
pre-commit + CI |
| nextest | cargo nextest run |
Cargo.toml |
CI |
| mutants | cargo mutants |
[workspace.metadata.cargo-mutants] in Cargo.toml |
CI weekly |
| arch_fitness | cargo test --test arch_fitness |
tests/arch_fitness.rs |
CI |
| insta snapshots | cargo insta review |
tests/behaviour_harness.rs |
CI |
| gitleaks | gitleaks detect |
.gitleaks.toml |
CI |
Steering Loop
When any sensor fires repeatedly (>2 times in one sprint):
- Identify the root cause category (maintainability / architecture / behaviour).
- Update the corresponding feedforward guide to prevent recurrence.
- If no guide exists, create one in
.agents/skills/using theskill-creatorskill. - Document the update in
CHANGELOG.md.
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 · 69 lines · 68 tokens per session scan A ed8e7b6d1797
harness is a skill published in the GitHub repository d-oit/rust-2026-template (10 stars, last pushed 3d ago), licensed MIT. It adds 68 tokens to every session and 765 once invoked, about $0.0003 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.
Other skills, from other repositories
NovaForge
NovaForge 通用知识笔记模板 — 将任意学科/项目的知识体系整理为结构化笔记。LaTeX+Typst+Markdown三版本，支持章节笔记（7步结构）、期末复习（真题分类+留白练习）、考研（7步+考研真题）、考公（行测/申论/面试）、科研（文献+笔记+方法）、项目（架构+进度+复盘）六种模式，适配考研/考公/专业课/科研/项目/竞赛。.
plumb-line-method
Use when a builder wants to learn or be reminded of the plumb-line method — the discipline of epistemic honesty enforced by tooling. Teaches the thesis, the nine portable principles, the maturity vocabulary, and the one-line test. Pure knowledge; takes no actions.
learn
Capture a learning (bug, workaround, pattern, gotcha) and save to persistent memory. Use when the user says "learn", "save this", "remember this", "note this", "record this", "never forget this", or when a non-obvious bug fix, workaround, or integration pattern is discovered during development.
capture-lessons
Ship and capture lessons workflow. Creates the PR, updates lessons learned, updates documentation, generates the sprint retrospective, and updates the backlog. Use at the end of a sprint cycle, or when the user says "ship", "capture", "lessons", "retro", "retrospective", "wrap up", or "done with this sprint". Also use…
full-sprint
Orchestrates a complete sprint cycle autonomously. Each phase runs as a separate claude -p session for context isolation. Use when the sprint-directive.md is ready and the sprint should execute end-to-end. Also use when the user says "run a full sprint", "execute the sprint", "build everything", "autonomous sprint"…
sprint-plan
Sprint planning workflow. Creates a structured plan from the backlog with tasks, acceptance criteria, dependencies, and risks. Use when starting a new sprint, planning work, or when the user says "plan", "sprint", "what should we work on", or "next sprint". Always use this skill to start a sprint cycle — never plan…