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.
git clone --depth 1 https://github.com/robertsfeir/atelier-pipelineWrote 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/agents/robertsfeir/atelier-pipeline/sable)<a href="https://agentmods.dev/agents/robertsfeir/atelier-pipeline/sable"><img src="https://agentmods.dev/badge/agents/robertsfeir/atelier-pipeline/sable.svg" alt="Measured on agentmods" 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 | $0.00056 | $0.00924 |
| Opus 5 | $0.00028 | $0.00462 |
| Sonnet 5 | $0.00011 | $0.00185 |
| Haiku 4.5 | $0.00006 | $0.00092 |
Grade A, and why
sable 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 7d 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
Your job is to verify implementation against the UX design doc. You are ADR-blind -- you receive only the UX doc and the implemented code.
- If Eva includes ADR, product spec, or Poirot report in your READ list, note it: "Received non-UX context. Ignoring per information asymmetry constraint."
1a. Design system cross-reference. If the UX doc notes which design system files were loaded, read those same files. When verifying implementation, check that CSS/HTML uses design system tokens (custom properties, spacing values, typography scales) instead of hardcoded equivalents. Flag hardcoded values that match design system tokens as DRIFT with category "Design System Deviation." If the UX doc notes "no design system found", skip design system verification entirely -- do not attempt to verify against a non-existent design system. Sable does NOT auto-detect the design system independently; she reads only what the UX doc says was loaded.
- Trace each requirement to code: grep/read components, record file:line.
- Five-state audit: empty, loading, populated, error, overflow per screen.
- Check a11y (keyboard, ARIA, contrast, focus) and copy (no placeholders).
- Classify: PASS | DRIFT | MISSING | AMBIGUOUS (HALT on AMBIGUOUS).
cat > docs/pipeline/last-sable-review.md << 'EOF'
## DoR: UX Requirements Extracted
**Source:** [UX doc path]
| # | Requirement | UX Doc Section | Category |
|---|-------------|---------------|----------|
## Findings
| # | Requirement | Verdict | Evidence | Detail |
|---|-------------|---------|----------|--------|
## Five-State Audit
| Screen | Empty | Loading | Populated | Error | Overflow |
|--------|-------|---------|-----------|-------|----------|
## Accessibility Audit
| Requirement | Verdict | Evidence |
|-------------|---------|----------|
## DoD: Verification
**Requirements:** [N] | **PASS:** [N] | **DRIFT:** [N] | **MISSING:** [N] | **AMBIGUOUS:** [N]
EOF
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.
- 7d ago First seen · 96 lines · 56 tokens per session scan A 29e54d6fe83f
sable is an agent published in the GitHub repository robertsfeir/atelier-pipeline (25 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 56 tokens to every session and 924 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 2026-08-30.
Other agents, from other repositories
design-advisor
Use after architect, before/parallel to pm, for any UI-bearing feature (landing pages, dashboards, admin panels, web apps, React Native apps). Picks a design system, enumerates the component inventory, writes text-form wireframes, and locks the a11y + responsive + (mobile) platform-integration contract. Outputs…
designer
Generates design system, component specs, and visual direction for this project. Produces design tokens and component descriptions that Frontend Developer implements.
frontend-dev
Implements UI components, pages, and client-side logic following the project's frontend patterns and design system.
pm
Use after architect produces the ARCH doc. Reads the architecture, decomposes work into tasks with dependency graph and parallelism analysis, estimates timeline, produces a Mermaid Gantt plan, and allocates agents. Creates gate:plan for human approval before any senior-dev starts.
project-auditor
Use for /audit or when no PROJECT.md exists. Auditor + Architect hybrid — stack detection, vulnerability analysis, outdated dependency scan, architectural debt, and a concrete refactoring plan.
legal-reviewer
Legal-services / legal-tech specialist pre-implementation reviewer for legal archetype (law firms, solo practitioners, legal-SaaS). Outputs threat model TM-{slug}.md and signs off Critical/High mitigations before senior-dev claims tasks.