feat: improve 5 lowest-scoring skill definitions

[rohan-tessl]

Hey @deepgram 👋

I ran your skills through tessl skill review at work and found targeted improvements in your skills. Here's the before/after:

Skill	Before	After	Change
deepgram-dotnet-management-api	89%	100%	+11%
deepgram-dotnet-speech-to-text	89%	100%	+11%
deepgram-dotnet-conversational-stt	76%	95%	+19%
deepgram-dotnet-text-to-speech	86%	95%	+9%
deepgram-dotnet-voice-agent	89%	95%	+6%

These were easy changes to bring the skill's structure and activation in line with what performs well against Anthropic's best practices. Added error handling, reconnection logic, cleanup/shutdown steps, condensed verbose sections.

In addition, I stress-tested your deepgram-dotnet-conversational-stt skill against a few real-world scenarios, and it held up really well. This means that your skill meaningfully improves agent steering and contributes to stronger output quality. Kudos for that!

Honest disclosure, I work at @tesslio where we build tooling around skills like these. Not a pitch, just saw room for improvement and wanted to contribute.

If you want to self-improve your skills, or define your own scenarios to pressure test, just ask your agent (Claude Code, Codex, etc.) to evaluate and optimize your skill with Tessl. Ping me @rohan-tessl, if you hit any snags.

[rohan-tessl]

Thanks @GregHolmes for the quick merge, and @lukeocodes for the review really appreciated the fast turnaround on a drive-by contribution.

We've reviewed a bunch of skills across different orgs by now and have picked up some useful patterns around keeping them maintainable, secure, and well-scored as adoption grows — especially around the tradeoffs between agent-friendly brevity and SDK-accurate detail (which mattered a lot for the conversational-stt and voice-agent skills here).

If you're open to it, a 15-min coffee would be a good way to swap notes: Calendly link.