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 skills add jxoesneon/Ciel --skill agent-systems-and-harnessesgit clone --depth 1 https://github.com/jxoesneon/CielWrote 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/jxoesneon/ciel/agent-systems-and-harnesses)<a href="https://agentmods.dev/skills/jxoesneon/ciel/agent-systems-and-harnesses"><img src="https://agentmods.dev/badge/skills/jxoesneon/ciel/agent-systems-and-harnesses/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/jxoesneon/ciel/agent-systems-and-harnesses"><img src="https://agentmods.dev/badge/skills/jxoesneon/ciel/agent-systems-and-harnesses.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 | $0.00028 | $0.00531 |
| Opus 5 | $0.00014 | $0.00266 |
| Sonnet 5 | $0.00006 | $0.00106 |
| Haiku 4.5 | $0.00003 | $0.00053 |
Grade A, and why
agent-systems-and-harnesses 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 10d 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.
What it actually says
CIEL ADAPTATION: Agent Systems (The Autonomy Layer)
This skill formalizes the operation of complex AI systems, from cloud-hosted workers to adversarial feedback loops.
Enterprise Agent Ops
- Lifecycle: HARD control over Start, Pause, Stop, and Restart. Hard timeouts on all turns.
- Observability: Track Success Rate, Cost per Task, and Failure Class Distribution.
- Least Privilege: Inject credentials via environment; PROHIBIT broad filesystem or network access without specific audit logs.
Adversarial Harnesses (GAN-Style)
- Generator: Implements features based on spec and evaluator feedback.
- Evaluator: Ruthlessly strict QA agent. Uses Playwright to test the LIVE app, not just code.
- Pass Threshold: Target weighted score > 7.0 across Design, Craft, and Functionality.
- Mandate: Evaluator MUST be a separate instance from Generator to prevent self-praising bias.
Headless Orchestration (GSD)
- Spec-Driven: Launch builds via
gsd headless new-milestone --context spec.md. - Polling: Use
queryfor instant status snapshots (no LLM cost). - Blocked State: If exit code is
10, query for blocker details and steer or escalate to human.
Anti-Patterns
- Self-Evaluation: Letting an agent approve its own PR or design.
- Amnesiac Deployment: Restarting an agent fleet without reading the previous session's persistent memory.
- Blind Auto-Build: Running an un-monitored GSD loop for > 30 minutes without a cost-limit cap.
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.
- 10d ago First seen · 51 lines · 28 tokens per session scan A 7ff313de57f8
agent-systems-and-harnesses is a skill published in the GitHub repository jxoesneon/Ciel (1 stars, last pushed 3d ago), licensed Apache-2.0. It adds 28 tokens to every session and 531 once invoked, about $0.0001 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
review-tests
Reviews test quality and coverage. Use when checking that a diff has adequate tests, checking a new or edited test, or auditing the test suite of a codebase.
testing-rate-limiting
Apply deterministic SlowAPI rate-limiter isolation for FastAPI transport tests, especially to avoid flaky HTTP 429 responses.
fluent
Operate the fluent workflow to build software autonomously over extended periods. Interactive stages (brief, behaviors, approach, plan) run with the user. Autonomous execution loops writer → tester → parallel reviewers, then runs the Learner after a passing review round. Only a successful Learner run produces a ready…
test-scaffold
Gera estrutura completa de testes automatizados (unitários, integração e edge cases) com base no código existente. Use para "escrever testes para este arquivo", "aumentar cobertura" ou "criar mocks".
e2e-runner
End-to-end testing specialist using Playwright. Use PROACTIVELY for generating, maintaining, and running E2E tests. Manages test journeys, quarantines flaky tests, uploads artifacts (screenshots, videos, traces), and ensures critical user flows work.
tdd-workflow
Use this skill when writing new features, fixing bugs, or refactoring code. Enforces test-driven development with 80%+ coverage including unit, integration, and E2E tests.