Connect
Minimal request shape, authentication checks, and first successful response.
Documentation Desk
Coding Agent Guide
Practical guides for AI coding Agent tutorials and operations.
agent-workflow.md
1
install tool and sandbox2
load repo context3
run tests4
prepare PR handoff122
Published guides
4
Implementation tracks
Source
Checked guidance
Documentation Map
AI coding agent guides by operating path
Pick the path that matches the workflow question: configure, route models, load context, or hand off a tested change.
Validate
Smoke tests, response contracts, and source-backed assumptions before traffic.
Harden
Fallback, observability, rate-limit behavior, and production handoff notes.
Review
Refresh decisions and editorial checks for pages that should stay useful.
Latest Docs
Recent implementation guides
Historical archive entries remain available to readers while staying out of RSS, sitemap, and llms.txt.
Gate Sensitive MCP Tool Calls With Human Approval in Coding Agent Runs
A practical design for pausing sensitive MCP actions, collecting clear approval, and recording sanitized evidence.
Prepare Bug Screenshots for Coding Agents Using CometAPI
A practical workflow for packaging, validating, and safely logging bug screenshots before coding agents analyze them through CometAPI.
Stop Coding Agents Without Leaving CometAPI Calls Running
Learn how to propagate a coding-agent stop signal into active CometAPI requests, streams, workers, and cleanup.
Catch Accessibility Regressions in Coding Agent UI Patches
Build a practical merge gate that combines automated accessibility scans, targeted human checks, and reviewable evidence.
Use OIDC to separate CometAPI model access from short-lived, claim-restricted cloud credentials in coding agent deployment jobs.
Separate model work from deployment authority by giving only a reviewed deployment job narrowly scoped, short-lived cloud credentials through OIDC.