Plain-language summary
- What it solves
- Present the lab's projects consistently without turning editorial automation into a source of new scientific results.
- Who it may help
- People following the lab, navigating its workflows or studying governed editorial automation.
- What it produces
- Scheduled bilingual posts, rotating spotlights, original cards and public engagement counters.
Make the portfolio readable and discussable through deterministic editorial scheduling without treating the workflow as a generator of scientific evidence.
3 · What it produces
IT/EN root posts with original cards for identity, spotlight, method, applications, roundup and community, plus state, deduplication and public engagement counters.
The previews come directly from the WF19 repository and represent the editorial format generated by the workflow.
4 · How it works
Deterministic editorial templates + registry-driven rotation + fail-closed publishing
5 · Data and signals
| Element | Detail |
|---|---|
| 5 · Data and signals | Canonical website workflow registry, JSON cache/fallback, campaign map, deterministic cycle key, Europe/Rome schedule and public Bluesky counters |
| Industrial transferability | technical communication, knowledge dissemination, portfolio discovery and governed campaigns |
| Research | controlled comparison across editorial families and public engagement measures |
| Interpretation condition | no language model in the writing path; claims are limited to documented workflows and publishing stops when gates cannot be verified |
deterministic selection of template, language, campaign and spotlight; no generative model in text composition
slot and blackout validation, weekly cap, deduplication, reviewable diff and fail-closed gates for schedule, policy and credentials
Python 3.11+, YAML/JSON, Pillow, GitHub Actions, AT Protocol/Bluesky
stale registry, unknown slot or unsupported claim; disclosed fallbacks and publication blocking reduce error propagation
Workflow technical dossier
Methods, checks, stack and transferability conditions.
5 · Data and signals
Canonical website workflow registry, JSON cache/fallback, campaign map, deterministic cycle key, Europe/Rome schedule and public Bluesky counters
Inference / calculation
deterministic selection of template, language, campaign and spotlight; no generative model in text composition
Validation
slot and blackout validation, weekly cap, deduplication, reviewable diff and fail-closed gates for schedule, policy and credentials
Stack
Python 3.11+, YAML/JSON, Pillow, GitHub Actions, AT Protocol/Bluesky
stale registry, unknown slot or unsupported claim; disclosed fallbacks and publication blocking reduce error propagation
See recent outputs
The Publications page collects local previews and, on request, the public feed.
Potential transferability
Have a problem similar to Lab Intelligence Community?
A first discussion can start from the objective, available data and success criterion without sending confidential information. A result observed here is not assumed to transfer automatically to another context.