{"@attributes":{"version":"2.0"},"channel":{"title":"TraceAxiom engineering field guides","link":"https:\/\/traceaxiom.com\/guides","description":"Practical writing on secure, verifiable AI coding-agent systems.","language":"en","lastBuildDate":"Mon, 05 Oct 2026 00:00:00 GMT","item":[{"title":"AI Coding Agents: From Code Generation to Accountable Delivery","link":"https:\/\/traceaxiom.com\/guides\/ai-coding-agents","guid":"https:\/\/traceaxiom.com\/guides\/ai-coding-agents","description":"Learn what separates AI coding agents from assistants, where they fail, and how contracts, independent gates, and human authority make them useful.","pubDate":"Fri, 02 Oct 2026 00:00:00 GMT"},{"title":"Secure AI Coding Agents: A Practical Control Model","link":"https:\/\/traceaxiom.com\/guides\/secure-ai-coding-agents","guid":"https:\/\/traceaxiom.com\/guides\/secure-ai-coding-agents","description":"Secure AI coding agents with least authority, isolated execution, deny-by-default egress, credential boundaries, source integrity, and verifiable receipts.","pubDate":"Fri, 02 Oct 2026 00:00:00 GMT"},{"title":"AI Code Verification: How to Check AI-Generated Pull Requests","link":"https:\/\/traceaxiom.com\/guides\/ai-code-verification","guid":"https:\/\/traceaxiom.com\/guides\/ai-code-verification","description":"Verify AI-generated code with candidate-bound build, behavior, accessibility, security, and acceptance gates that can fail independently.","pubDate":"Fri, 02 Oct 2026 00:00:00 GMT"},{"title":"Coding Agent Evaluation That Measures Accepted Outcomes","link":"https:\/\/traceaxiom.com\/guides\/coding-agent-evaluation","guid":"https:\/\/traceaxiom.com\/guides\/coding-agent-evaluation","description":"Evaluate coding agents with locked tasks, independent acceptance, provider-reported usage, reproducible evidence, and honest sample limitations.","pubDate":"Fri, 02 Oct 2026 00:00:00 GMT"},{"title":"How to Verify a Coding Agent\u2019s Runtime Evidence","link":"https:\/\/traceaxiom.com\/guides\/verify-agent-runtime-evidence","guid":"https:\/\/traceaxiom.com\/guides\/verify-agent-runtime-evidence","description":"Review a coding agent\u2019s runtime evidence: identify the observed process, verify signed package bytes, inspect missing coverage, and keep execution approval separate.","pubDate":"Fri, 02 Oct 2026 00:00:00 GMT"},{"title":"How to Calculate Cost per Accepted Coding-Agent Change","link":"https:\/\/traceaxiom.com\/guides\/cost-per-accepted-agent-change","guid":"https:\/\/traceaxiom.com\/guides\/cost-per-accepted-agent-change","description":"A worked example of coding-agent cost per accepted change: include failed attempts, verification and review, separate credits, and handle missing acceptance.","pubDate":"Fri, 02 Oct 2026 00:00:00 GMT"},{"title":"How to Check CI Artifacts Against the Exact Code Under Review","link":"https:\/\/traceaxiom.com\/guides\/verify-ci-artifacts-for-exact-code","guid":"https:\/\/traceaxiom.com\/guides\/verify-ci-artifacts-for-exact-code","description":"A worked CI artifact review: match the source, workflow and attempt, check original bytes, preserve mismatches, and separate successful checks from acceptance.","pubDate":"Fri, 02 Oct 2026 00:00:00 GMT"},{"title":"How to Verify Cleanup After a Coding-Agent Sandbox Run","link":"https:\/\/traceaxiom.com\/guides\/verify-coding-agent-sandbox-cleanup","guid":"https:\/\/traceaxiom.com\/guides\/verify-coding-agent-sandbox-cleanup","description":"A worked sandbox cleanup review: retain results, confirm owned resources and runner removal, preserve failed observations, and reconcile usage separately.","pubDate":"Fri, 02 Oct 2026 00:00:00 GMT"},{"title":"Measure Coding-Agent Speedup","link":"https:\/\/traceaxiom.com\/guides\/measure-coding-agent-speedup","guid":"https:\/\/traceaxiom.com\/guides\/measure-coding-agent-speedup","description":"A worked example of coding-agent speedup that includes integration, verification, review and retries, with explicit treatment of failed and unfinished tasks.","pubDate":"Fri, 02 Oct 2026 00:00:00 GMT"},{"title":"Review MCP Tool Permissions","link":"https:\/\/traceaxiom.com\/guides\/review-mcp-tool-permissions","guid":"https:\/\/traceaxiom.com\/guides\/review-mcp-tool-permissions","description":"A worked MCP permission review: match the tool contract, bound the grant, reject invalid requests and schema drift, and keep returned content untrusted.","pubDate":"Fri, 02 Oct 2026 00:00:00 GMT"}]}}