An Agent Skill that builds a skill with evals out of the conversation in which the agent finally got it right.
100%
Does it follow best practices?
Impact
100%
3.84xAverage score across 3 eval scenarios
Passed
No findings from the security scan
Changing AGENTS.md or files under docs/agents/.
AGENTS.md holds only the one-sentence project description, the startup instruction, the
durable-correction rule, and links. Topic guidance goes in a page under docs/agents/, linked
from AGENTS.md.AGENTS.md, write it into the matching docs/agents/ page, or
a new page from the template below, and add only a link to AGENTS.md.docs/agents/ must make sense on their own. Summarize a decision and its reason
instead of pointing at an issue, a pull request, a transcript, or a hosted run. The validator
rejects issue numbers, pull request numbers, and run IDs here.# <Title>
## Scope
What this page covers.
## Rules
- Clear, actionable rules.
## References
- Related pages..tessl-plugin
skills
build-a-skill
assets
template
.tessl-plugin
evals
01-settings-loader-errors
evals-reference
01-webhook-error-review
evals-regression
02-hard-stop-audit
scripts
skills
my-skill
references