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-kovac-deterministic-memory/references/

Source outline

Source: Veljko Kovac, Memory Without Deterministic Coordinates Is Just Hearsay, AI Native DevCon New York 2026. Recording. All anchors refer to original numbered cues in the transcript. Short quotations below preserve the supplied transcription, including spoken wording.

TopicSource anchorExact short quoteApplication and scope
Sharing and permissions00:01:59–00:02:34, cues 35–46“Which agent can share information with what other agent?”Cross-agent memory requires access boundaries.
Conflict example00:03:33–00:05:34, cues 62–93“same function ID version allows you the chance to compare side by side.”Coordinates allow comparison without deciding which claim is true.
Ambiguous finding00:06:32–00:08:33, cues 111–142“client send and you had async client send.”A useful-sounding learning can be attached to the wrong declaration.
Shared files and instructions00:09:48–00:11:46, cues 162–196“a pile of MD files.”Shared text alone does not resolve attribution or lifecycle.
Skill activation concern00:11:51–00:13:18, cues 197–218“who picks the skill, right.”Availability does not guarantee retrieval or correct attribution.
Fact and snapshot00:14:02–00:16:31, cues 230–269“facts should be looked up, not decided.”A repository snapshot alone does not identify the relevant function.
Coordinate requirements00:16:53–00:19:22, cues 275–313“it should be stamped by a tool.”The talk requires deterministic tool-produced identity and version awareness.
Connected graph00:19:26–00:21:03, cues 314–337“group by operation on the ID”Stable keys enable aggregation across findings.
Dependency change00:21:06–00:21:26, cues 338–343“some other helper that was connected with that function could have changed.”Unchanged function text does not imply unchanged applicability.
Reported experiment00:21:30–00:23:47, cues 344–377“we did run this five times.”The comparison concerns attribution and grouping on a small example.
Organizational learning00:23:50–00:26:21, cues 378–420“every single conflict that is being resolved, it's being done for a reason.”Resolved findings can inform organizational memory with provenance.
Retention Q&A00:26:36–00:30:41, cues 424–486“there is no golden answer.”Retention and forgetting depend on the use case and ongoing evaluation.

Source limitations

  • The source does not specify the exact identifier algorithm, parser versions, collision handling, rename semantics, or complete dependency analysis.
  • Stable coordinates improve attribution and joins; they do not prove a finding's truth, applicability, completeness, or permission to share.
  • The small experiment and third-party skill benchmarks are reported claims, not independently validated universal performance evidence.
  • Harness memory-loading behavior and product capabilities are talk-time descriptions. The Q&A's keep-latest simplification must be read with the earlier warning that latest need not mean correct.
  • Promotional interlude cues 219–229 are omitted from the bundled transcript; the remaining cue numbers retain their source positions.
  • This is a spoken transcript; unseen slides, linked repositories, and third-party research were not independently inspected. No outcome beyond the described talk is established.

skills

talk-kovac-deterministic-memory

tile.json