Practical patterns for building governed AI agents, platform architecture, and distributed systems — from real-world engineering experience.https://allsrc.dev/https://allsrc.dev/posts/pattern-12-agent-lifecycle-profile/https://allsrc.dev/posts/pattern-12-agent-lifecycle-profile/A pattern for treating agents as governed runtime assets with identity, owner, purpose, limits, and retirement state.Mon, 29 Jun 2026 10:00:00 GMThttps://allsrc.dev/posts/pattern-11-cicd-evaluation-gates/https://allsrc.dev/posts/pattern-11-cicd-evaluation-gates/A pattern for blocking unsafe agent changes before they reach production.Wed, 24 Jun 2026 10:00:00 GMThttps://allsrc.dev/posts/pattern-10-agent-evaluations/https://allsrc.dev/posts/pattern-10-agent-evaluations/A pattern for evaluating agent behavior through trajectories, policy checks, groundedness, and cost limits.Fri, 19 Jun 2026 10:00:00 GMThttps://allsrc.dev/posts/pattern-09-sandboxed-execution/https://allsrc.dev/posts/pattern-09-sandboxed-execution/A pattern for constraining agent-directed code execution with workspace, command, network, and approval policies.Sun, 14 Jun 2026 10:00:00 GMThttps://allsrc.dev/posts/pattern-08-memory-isolation/https://allsrc.dev/posts/pattern-08-memory-isolation/A pattern for making agent memory scoped, typed, validated, and isolated across users and agents.Tue, 09 Jun 2026 10:00:00 GMThttps://allsrc.dev/posts/pattern-07-rag-access-control/https://allsrc.dev/posts/pattern-07-rag-access-control/A pattern for enforcing authorization during retrieval and carrying source provenance into generated answers.Thu, 04 Jun 2026 10:00:00 GMThttps://allsrc.dev/posts/pattern-06-prompt-injection-goal-hijack/https://allsrc.dev/posts/pattern-06-prompt-injection-goal-hijack/A pattern for treating untrusted content as data and preserving the original user goal.Sat, 30 May 2026 10:00:00 GMThttps://allsrc.dev/posts/pattern-05-redaction-boundary/https://allsrc.dev/posts/pattern-05-redaction-boundary/A pattern for sanitizing sensitive data before it crosses model, tool, display, or audit boundaries.Mon, 25 May 2026 10:00:00 GMThttps://allsrc.dev/posts/pattern-04-tool-budgeting/https://allsrc.dev/posts/pattern-04-tool-budgeting/A pattern for bounding model calls, tool calls, retries, tokens, handoffs, and per-run spend.Wed, 20 May 2026 10:00:00 GMThttps://allsrc.dev/posts/pattern-03-decision-trace/https://allsrc.dev/posts/pattern-03-decision-trace/A pattern for reconstructing agent decisions through structured, redacted events without storing private reasoning.Fri, 15 May 2026 10:00:00 GMThttps://allsrc.dev/posts/pattern-02-hitl-approval-gate/https://allsrc.dev/posts/pattern-02-hitl-approval-gate/A pattern for placing human review only where agent actions need judgment, accountability, or reversibility checks.Sun, 10 May 2026 10:00:00 GMThttps://allsrc.dev/posts/pattern-01-tool-privilege-broker/https://allsrc.dev/posts/pattern-01-tool-privilege-broker/The first pattern in our Agent Harness series: defining a deterministic policy boundary between the model's reasoning and the execution of external tools.Wed, 06 May 2026 10:00:00 GMThttps://allsrc.dev/posts/what-is-an-agent-harness/https://allsrc.dev/posts/what-is-an-agent-harness/An introduction to the Agent Harness series: defining why the LLM is just a reasoning engine and how the harness manages memory, tools, and execution boundaries.Fri, 01 May 2026 10:00:00 GMT