Open-source, secure environment with real-world tools for enterprise-grade agents.
Python agentaiai-agentai-agentscode-interpretercopilotdevelopmentdevtoolsgptgpt-4javascriptllmnextjsopenaipythonreactsoftwaretypescript
Update smol deploy todo list
Tomas Valenta095030bb239e
diff --git a/SMOL_DEPLOYMENT.md b/SMOL_DEPLOYMENT.mdindex 16159acae..1c5049f4c 100644--- a/SMOL_DEPLOYMENT.md+++ b/SMOL_DEPLOYMENT.md@@ -6,11 +6,15 @@- They can also modify permissions for our GH App if we don't have access to repo they want to use- User adds initial prompt that specifies what should smol dev do in the first iteration- User specifies their OpenAI key and clicks on "Deploy"-- Then we create a PR throught the GH API+- Then we create a PR throught the GH API (using the GH App token)+ - We need to create branch and (empty) commit first to be able to create PR+ - (For now we can probably create the branch directly in the repo without forking it - we need repo write access for that)- We then create agent deployment for the specific repo- The necessary GH data are in the `auth` field in the deployments table- The OpenAI key is included in the `config` field alongside the model configuration- We also add informations about the PR that was created so we can update the branch connected to the deployment via git pushes in the agent dev environemnt- During each agent run we authenticate with GH App token, pull the repo, make the changes necessary and then push the changes to the branch connected to the deployment+ - To pull and push we use the `git` command line tool but we need to authenticate with GH App token+ - We can use the `git credential` helper to do that - ideally using the `gh` CLI tool- On each comment to the PR we trigger a webhook that compiles the prompt from the PR description and comments and then triggers a new run of the smol agent- If the smol dev agent is already processing previous PR change we cancel that run and then start a new one with the latest changes