Skip to content

Realign docs and engine contracts to the steering-first direction - #16

Merged
nimser merged 1 commit into
mainfrom
feat/new-direction-alignment
Jul 5, 2026
Merged

Realign docs and engine contracts to the steering-first direction#16
nimser merged 1 commit into
mainfrom
feat/new-direction-alignment

Conversation

@nimser

@nimser nimser commented Jul 5, 2026

Copy link
Copy Markdown
Owner

Part 1 of 2 (part 2: #15, stacked on this branch). Pure alignment to the new direction — docs, ADRs, and contract-level code changes only; no new features.

ADRs

  • New ADR-009 — adapter scope and tiering: Tier 1 = Pi + Claude Code (first-party); everything else community-owned against the stable interface
  • New ADR-010user-input phase (design; implementation ships in feat: steering-first surface — change-17 user-input phase, change-18 QoL steering packs, change-7 claude-code adapter #15)
  • ADR-002 rewritten — three phases, Tier 1 capability table (incl. redactUserInput, Claude Code redact floor ≥ 2.1.121)
  • ADR-003 rewritten — minimal public API: createEngine(packs, capabilities)evaluate(ctx), pack loader, public types; everything else @internal
  • ADR-008 rewritten — new non-goals: no policy-as-code engine, not a security boundary (agentjail named as the containment companion)
  • ADR-001/005/006/007 swept for dropped-adapter references and historical asides

Public docs

  • README rewritten: steering-scenario hero, batteries included, the three doors, steering/security pack split (--strict stated as roadmap, per ADR-007)
  • New LIMITATIONS.md; SECURITY.md links it and answers never-inject vs. redact
  • getting-started, rule-pack-guide, CONTRIBUTING updated to the Tier 1 world

Code (contract alignment only)

200 tests green; lint/format/typecheck pass.

🤖 Generated with Claude Code

https://claude.ai/code/session_01FUpa7w5hYzrXyCYkK8K1DC

…tion

ADR set rewritten around the new direction: ADR-009 (Pi + Claude Code
as the only Tier 1 adapters), ADR-010 (user-input phase, spec only),
ADR-003 (minimal createEngine()/evaluate() embeddable API), ADR-008
(no policy-as-code, not a security boundary), ADR-002 (three phases,
Tier 1 capability table). README rewritten (steering hero, three
doors, honest pack status), new LIMITATIONS.md, SECURITY.md pairing
answer. Code alignment only: createEngine facade, internals marked
@internal, Tier 1 capability constants, user-input types/validation
plumbing (engine wiring lands with the feature PR), and the
confirm→block fallback per ADR-002 (was confirm→suggest).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FUpa7w5hYzrXyCYkK8K1DC
@coderabbitai

coderabbitai Bot commented Jul 5, 2026

Copy link
Copy Markdown

Warning

Review limit reached

@nimser, you've reached your PR review limit, so we couldn't start this review.

Next review available in: 28 minutes

Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available.
You're only billed for reviews past your plan's rate limits ($0.25/file).

How can I continue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews.

How do review limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please refer docs for additional details.

Review details
⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: ASSERTIVE

Plan: Pro Plus

Run ID: 315f1d4d-3e83-43c0-96fa-913fd5426fa1

📥 Commits

Reviewing files that changed from the base of the PR and between 781ae79 and b21d72c.

📒 Files selected for processing (24)
  • .gitignore
  • CONTRIBUTING.md
  • LIMITATIONS.md
  • README.md
  • SECURITY.md
  • docs/adrs/001-layered-architecture.md
  • docs/adrs/002-behavior-model.md
  • docs/adrs/003-public-api-contract.md
  • docs/adrs/005-yaml-rule-packs.md
  • docs/adrs/006-observability-strategy.md
  • docs/adrs/007-trust-and-self-protection.md
  • docs/adrs/008-non-goals.md
  • docs/adrs/009-adapter-scope-and-tiering.md
  • docs/adrs/010-user-input-mediation.md
  • docs/getting-started.md
  • docs/rule-pack-guide.md
  • src/core/harness-capabilities.ts
  • src/core/normalizer.ts
  • src/core/types.ts
  • src/core/validator.ts
  • src/engine/create-engine.ts
  • src/index.ts
  • src/resolver/action-resolver.test.ts
  • src/resolver/action-resolver.ts
✨ Finishing Touches
✨ Simplify code
  • Create PR with simplified code
  • Commit simplified code in branch feat/new-direction-alignment

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

Comment @coderabbitai help to get the list of available commands.

@sonarqubecloud

sonarqubecloud Bot commented Jul 5, 2026

Copy link
Copy Markdown

Quality Gate Failed Quality Gate failed

Failed conditions
79.8% Coverage on New Code (required ≥ 80%)

See analysis details on SonarQube Cloud

@nimser
nimser merged commit 9861b48 into main Jul 5, 2026
4 of 6 checks passed
@nimser
nimser deleted the feat/new-direction-alignment branch July 5, 2026 20:55
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant