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-cox-secure-agentic-workflows/references/

Source outline

Use transcript.md for exact wording. Promotion cues 227–237 are omitted.

TimeCuesTopic
00:00–02:011–28Background and event-time release account
02:07–03:4429–50Actions platform, Markdown and deterministic compilation
03:49–06:3451–88Maintenance automation and small changes
06:38–10:2689–147Team repository observations and debt cleanup
10:34–13:26148–188Secrets, read filtering and safe outputs
13:29–14:31189–201Isolation, proxies and credential boundaries
14:37–16:21202–226Logs, optimization and buffered writes
17:09–17:57238–249Post-agent deterministic output filtering
18:05–20:04250–277Cross-repository automation direction
20:09–22:18278–311Hypothetical public/private oversharing
22:24–26:10312–362Primary agent, enclave and trusted broker
26:14–28:33363–396Schema width, budgets and new-feature caveat

Short quotations

  • “our agents run in a read only mode.” — excerpt, cue 173, 12:17.
  • “The primary never touches private data.” — excerpt, cue 316, 22:42.
  • “It's not it's not real.” — excerpt, cue 305, 21:53; the hypothetical example.

Limits

The transcript has no Q&A. Integrity filtering is named but not explained fully. Isolation claims assume the agent remains within its boundary. Safe outputs enforce configured policy, not complete correctness. The enclave budget description does not establish cumulative-query or side-channel guarantees. The private-launch example is explicitly fictional, not a speaker confidentiality restriction.

skills

talk-cox-secure-agentic-workflows

tile.json