AI Native DevCon 2026 New York — practical agent skills grounded in conference talks
86%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Low
Low-risk findings worth noting
This worksheet is a recommended application of Fahmy's design lessons. It is not executable SDK documentation.
Task: Describe the tool being built and what must happen deterministically.
Model judgment: Identify where an agent may choose an approach and where every input must be processed.
Coverage record: State how scheduled, completed, failed and unprocessed items are distinguished. Invoking a harness does not prove its analysis is correct.
Harness: List required invocation, progress, steering, interruption, completion and error behavior.
Workspace: List files, history, settings, skills and integrations needed for the task. Separate portable content from machine-specific configuration and credential references.
For each operation, name the owner and the evidence that it succeeded. Leave API names conceptual until the actual library is inspected.
| Requirement | Harness A | Harness B | Unsupported-case behavior |
|---|---|---|---|
| Fill from the workflow | Verified, unknown or absent | Verified, unknown or absent | Explicit fallback or stop |
Include child-process cleanup, approval semantics and error channels if relevant. Preserve meaningful differences rather than hiding them in a generic success string.
Record the source and destination environment, files copied, history exported, configuration restored and state deliberately omitted.
State that message-history portability is different from transferring a model's exact effective context or hidden reasoning. Identify missing dependencies before calling a session resumable.
Propose one representative local-to-remote task and one failure case. Check output integrity, coverage, observable errors, cleanup and permission boundaries.
Decision: Adopt the minimum surface needed, list unsupported scenarios and define when broader abstraction would be justified. Mark all tests as planned until performed; do not claim universal harness compatibility from a single successful handoff.
.tessl-plugin
skills
talk-arcolano-autonomy-and-trust
talk-aroussi-decision-context
talk-bevin-agent-assisted-instrument-design
talk-boeckeler-context-lifecycles
talk-braun-context-graphs
talk-carmont-evolving-agent-harnesses
talk-chanezon-gordon-agent-experience-evaluation
talk-cox-secure-agentic-workflows
talk-cunniffe-reusable-agent-platform
talk-davis-agent-tool-policies
talk-debois-concurrent-delivery
talk-dzialets-team-automation
talk-fahmy-harness-api-design
talk-finster-continuous-delivery
talk-fowler-regenerative-software
talk-gavrilescu-factory-architecture
talk-graciolli-amin-verification-aware-programming
talk-gruenbaum-reverse-graph-engineering
talk-hassan-agentic-engineering
talk-hossain-enterprise-agent-adoption
talk-huntley-patch-ready-factories
talk-jain-verifying-intent
talk-katsioloudes-montei-agentic-lifecycle
talk-knox-harness-engineering
talk-kovac-deterministic-memory
talk-lee-perkins-code-readiness
talk-lorenc-machine-speed-security
talk-lubken-software-of-one
talk-meijer-software-specialization
talk-meppiel-versioned-agent-contracts
talk-moss-concurrent-agent-risk
talk-myers-understanding-agent-generated-code
talk-ocytko-scaling-agentic-engineering
talk-oza-automating-design-quality
talk-panzica-agentic-sdlc
talk-pari-patel-agent-awareness
talk-podjarny-trustworthy-software-factories
talk-poulter-factory-improvement
talk-pratik-patel-refined-rag-data
talk-re-ferre-development-platforms
talk-sadogursky-factory-of-factories
talk-shah-training-agent-skills
talk-sharma-mohapatra-evolving-intent
talk-singhal-operating-agents-at-scale
talk-stack-incident-dossiers
talk-tyler-team-coding-benchmarks
talk-wotherspoon-agent-execution-environments
talk-yegge-ai-powered-factories
talk-zuber-leadership-through-coding