Automation

Checks and maintenance run mechanically instead of being remembered each time.

39 items in this guidance cluster.

Tagged guidance

Make bad output hardTurn recurring bad agent output into scripts, templates, lint, or checks that fail fast. Mechanical enforcement lowers review cost more reliably than repeated prompt reminders.Prefer tools over promptsMove repeated instructions into tools, checks, templates, or durable guides. Tooling catches failures even when prompts are short, compacted, or interpreted differently.Turn feedback into guidanceConvert repeated review corrections into rules, templates, snippets, or checks. Durable guidance fixes the workflow problem instead of requiring the same steering again.Verify risky changes with canariesUse staged rollout, shadowing, dry runs, or partial publication for changes that can fail only under real conditions. Canaries reduce blast radius while evidence accumulates.

Make exec tools noninteractiveDefault agent, CI, and background exec paths away from prompts, editors, pagers, and credential UI. Commands then fail or complete predictably, while human interactive modes remain explicit opt-ins.Stage generated behaviorValidate generated, reloadable, or plugin-provided behavior in a staging path before promoting it over known-good behavior. Staging adds recovery cost only where runtime-loaded or generated artifacts can fail after deployment.

Respect generated artifact ownershipFix generated outputs through their source inputs, templates, metadata, or generator configuration whenever those own the result. Hand edits are durable only when the tool workflow explicitly supports curation.Sync generated artifactsUpdate checked-in generated outputs when their source inputs change and they are part of review. Keeping them aligned prevents consumers from seeing stale artifacts.

Run timing benchmarks sequentiallySerialize timing-sensitive benchmarks when their numbers will be used as evidence. Concurrent runs compete for shared resources and can make the comparison describe the runner more than the code.

Deny accidental unsafeUse a crate-level lint when a crate intends to avoid unsafe code entirely. Executable policy catches accidental unsafe before it becomes normal implementation detail.Encode durable rules in lintsUse lint configuration for project policies stable enough to automate. Durable lints catch repeated mistakes without turning subjective taste into CI noise.Keep CI high signalKeep required Rust checks strict, fast, deterministic, and actionable. Slow or flaky ritual trains maintainers to ignore failures while real drift accumulates.Keep lints actionableEnforce lints that improve correctness, API quality, docs, portability, or maintenance in ways reviewers want automated. Scope suppressions tightly so exceptions stay visible.Prefer expect for lint suppressionsUse #[expect] for targeted lint suppressions that should disappear when the warning is fixed. Reserve broad allow attributes for deliberate policy choices that are not expected to expire.

Check maintainer commands in CIPut documented maintainer commands, or intentionally stronger equivalents, in CI. This keeps local instructions honest while leaving slow or credentialed checks to special jobs.Check MSRV and platformsRun checks for declared Rust versions and supported platforms when they are promised. Compatibility claims are public contracts, so skip matrices only when no such claim exists.Keep slow checks out of PR CIKeep long fuzzing, exhaustive matrices, and noisy benchmarks out of required PR CI. Fast gates preserve review flow, while heavier checks belong in release or scheduled validation.Run docs as a first-class gateTreat documentation checks as real validation for examples, links, commands, and claims. Prose-only edits may need less proof, but API-facing docs should fail before stale guidance ships.Run fast format and lint gates earlyRun cheap format and lint gates early in the feedback loop. They remove mechanical failures quickly, but they do not replace validation of risky behavior.

Avoid interactive JJ in agent workUse noninteractive jj commands with explicit messages, targets, and files in agent work. Interactive editors, prompts, merge tools, and pagers can hang unattended sessions.Quote revsets and shell syntaxQuote jj revsets, bookmark syntax, and other shell-sensitive command fragments. Shell metacharacters can alter commands, so examples should prefer simple safe quoting.Run JJ mutations sequentiallyDo not overlap jj commands that write repo, working-copy, bookmark, or config state. Sequential mutations avoid locks, stale reads, and confusing operation order.

Code is memory of processCode accumulates the history of decisions, shortcuts, and workflows that produced it. When changing it, preserve useful operational memory while removing obsolete process scars that no longer serve readers.Garbage collect agent driftAgent-generated notes, plans, and assumptions can outlive their usefulness and mislead later work. Periodically remove, update, or promote them so durable guidance stays current and temporary scaffolding does not become false context.Keep automations repo ownedCritical checks and generation steps become fragile when they live only in personal scripts or local memory. Put repeatable automation in the repository so agents and humans can run the same workflow.Make bad output mechanically hardRepeated review failures usually belong in a tool, type, schema, template, lint, or checklist instead of another reminder. Encode stable rejection criteria early, but keep escape hatches when legitimate exceptions are common.Practice agent workoutsHigh-risk or repeated agent workflows are unreliable when first exercised during real work. Build safe practice tasks with explicit win conditions so the workflow, skill, or runbook can improve before pressure arrives.Prefer tools over promptsProse instructions are weak guardrails for repeatable constraints because agents can forget or apply them too late. Use tools, tests, templates, and checks for mechanical preferences, while reserving prompts for judgment and tradeoffs.Teach agents through toolsRepeated agent confusion often means the repo lacks a command, fixture, wrapper, or check rather than another instruction. Convert repeatable setup, inspection, validation, and remediation into tools when they will improve future runs.Turn feedback into guidanceRepeated steering can expose missing, unclear, unreached, or ignored guidance. Diagnose the failure, fix the current artifact, and repair the instruction or delivery mechanism that should prevent recurrence.Verify with canaries before cutoverAutomation is risky when production cutover is the first real proof. Use a canary path, rollback plan, and human approval for operations with remote access, credentials, production data, or long-lived service impact.

Agent instructions are operational controlsAgent instructions should change future behavior, not merely restate a preference. Compact rules need enough trigger, purpose, and boundary context for agents to apply them without blind literalism.Mechanize repeated feedbackRepeated corrections should become checks, templates, generators, lints, or reusable guidance when the pattern is stable. Mechanizing the feedback moves human attention back to judgment instead of recall.

Guidance content model stewardshipContent-model stewardship keeps guides, rules, patterns, principles, mechanisms, references, snippets, and downstream templates in distinct roles. The mechanism combines mechanical inventory checks with maintainer review queues for overlap, thin guidance, noisy tags, and subjective content-quality decisions.Guidance generation and auditGenerators and audits keep rule indexes, compressed agent snippets, downstream templates, and Markdown synchronized. The checks catch recurring drift so review attention stays on judgment instead of mechanical bookkeeping.Rust lints and formattingRust and Markdown formatting plus targeted project lints encode stable mechanical preferences. The mechanism removes repeat style debate from review while leaving design judgment to humans and higher-level guidance.Rust tooling profileThe broad Rust tooling profile gathers lints, docs, release, CI, performance, and agent workflow checks in one place. Use it as a menu of mechanical supports, not as a substitute for project-specific risk judgment.Testing and benchmarkingTest, feature-matrix, fuzzing, property, and benchmark tools should be chosen by risk and feedback cost. This mechanism helps match validation depth to the behavior, integration surface, or performance claim under review.

Reviewed rule agent packGenerated compact execution surface for every reviewed rule. The pack gives agents stable rule IDs and instructions while leaving rationale, examples, and deeper context in the canonical rule files.