CtrlK
BlogDocsLog inGet started
Tessl Logo
PluginPublic

ainativedev/aidevcon-2026-nyc

AI Native DevCon 2026 New York — practical agent skills grounded in conference talks

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

Overview
Quality
Evals
Security
Files

outline.mdskills/talk-aroussi-decision-context/references/

Source outline

Read transcript.md for exact evidence. Promotion cues 157–167 are omitted.

TimeCuesTopic
00:00–01:091–14Background and evidence references
01:13–03:0515–40Net 30 versus missing Net 60 negotiation
03:09–04:3441–59Incomplete coverage, agreement status and authority
04:40–06:3460–85Scale and organizational ownership argument
06:40–09:2986–122Conversation memory versus organizational model
09:36–12:21123–156Retrieval critique and provenance-backed records
13:07–15:15168–196Distillation and traceable claims
15:21–18:02197–228Reported benchmark, ingestion and quarantine results
18:08–20:39229–260Permissions, history and coverage-aware unknowns
20:45–22:59261–291Aggregate-score limitations and contradiction ranking
23:04–27:13292–344Authorization before model context and attack tests
27:16–30:24345–382Local/hybrid results and takeaways

Short quotations

  • “A prompt is not a gate.” — excerpt, cue 293, 23:07.
  • “we can not conjure it into existence.” — excerpt, cue 249, 19:41.
  • “Keep the history because you need that, chain of evidence.” — cue 372, 29:32.

Limits

Benchmark results are speaker-reported and use differing tasks/denominators. The ingestion claim count includes a spoken correction. Lossy distillation is not proved lossless in general. Contradiction ranking remains a problem. External-model knowledge-leakage statements are the speaker's speculation, not verified provider behavior. Broad agent connectivity is described as upcoming. No Q&A is captured.

skills

talk-aroussi-decision-context

tile.json