test(reborn): freeze the LocalDev* type-name ratchet (§4.4/§10)

[ilblackdragon]

§4.4/§10 ratchet — freeze the LocalDev* type-name inventory

Stacked on #6204. Adds the deployment-mode-as-type ratchet §4.4/§10 of docs/reborn/2026-07-17-architecture-simplification-dto-dyn-local.md calls for, ahead of Slice B (collapsing the LocalDev* shadow runtime to a DeploymentConfig value).

§4.4's rule: a deployment mode is a config value, never a type the kernel or a substrate names. Today a whole LocalDev* type family encodes local-dev as types (approval/capability/lease/mount/network/outbound policy, the store aliases, the root filesystem, the turn-state store, the synthetic-capability + extension-surface families). This test freezes that inventory and enforces the direction.

What it does

crates/ironclaw_architecture/tests/reborn_localdev_typename_ratchet.rs scans crates/ for pub/pub(crate) LocalDev* type definitions (struct/enum/trait/type alias) and asserts the set exactly equals a frozen allowlist (the current 31):

• a new LocalDev* type → fails: resolve the mode to policy data at the composition edge, don't grow another type.
• deleting one without trimming the allowlist → fails: the list shrinks in lock-step as Slice B lands (§10: compare set membership, never a count).

Definition of done for this axis: the allowlist reaches empty — local-dev is one DeploymentConfig constant, no LocalDev* type remains.

Notes

• Scoped to LocalDev* (clean empty-set goal). The broader §4.4 name audit — Bucket 2 renames (LocalFilesystem→DiskFilesystem, LocalHostProcessPort→HostProcessPort) and Bucket 3 false positives (Locale, HostedMcp* = hosted MCP endpoint not deployment mode, LocalTraceSubmission*) — is a separate concern, deliberately not folded in so this ratchet stays high-signal.
• The scanner is exhaustive: it surfaced 9 LocalDev* types a line-oriented grep had missed (synthetic-capability, extension-surface, auth-read-model, capability-wiring families) — exactly the §4.4.1 category-3 "genuine local-only mechanism" cluster Slice B must promote to first-party capabilities.

Safety

Test-only, ships with a scanner self-test. No production code touched.

🤖 Generated with Claude Code

---

Update (review pass)

ee85b4ca8 consolidates both §10 ratchets onto a shared hardened scanner (tests/ratchet_support/mod.rs) — comment/string stripping, restricted-visibility + unsafe/auto matching, occurrence-preserving scans, production-scoped walk — so this PR now also refactors reborn_inmemory_store_ratchet.rs onto that module (behavior identical, self-tests preserved). The LocalDev ratchet additionally gains cfg-aware multiplicity: factory's durable/no-durable #[cfg]-gated alias pairs (incl. multi-line gates) are exempt from the duplicate check only when every occurrence is gated; mixed pairs still fail. Regression-tested for all of the above.

[gemini-code-assist]

Caution

The consumer version of Gemini Code Assist on GitHub has been sunset. All code review activity has officially ceased.

[ironloopai]

🔎 IronLoop Review Status

Head: 12692536057e807c59c51ee4a0187321f5b5b60f
Result: One or more review results were superseded by a newer PR head.
Next: Run @ironloopai review on the latest PR head.
Updated: 2026-07-17T23:19:38.095Z

Current reviewers:

Reviewer	State	Verdict	Findings	Last update
ironloop/common-reviewer (reviewer)	Superseded	N/A	N/A	2026-07-17T20:29:53.628Z

Reviewer summaries

Reviewer	Detail
ironloop/common-reviewer (reviewer)	Superseded by a newer PR head. New head: a6061d9. Previous verdict: Changes requested.

Recent activity

Time	Reviewer	State	Detail
2026-07-17T20:19:27.505Z	ironloop/common-reviewer (reviewer)	Queued	Accepted review request for head 7a2d4e5.
2026-07-17T20:19:27.505Z	ironloop/common-reviewer (reviewer)	Queued	Waiting for this reviewer lane to become available.
2026-07-17T20:19:28.294Z	ironloop/common-reviewer (reviewer)	Started	Reviewer worker started.
2026-07-17T20:19:30.876Z	ironloop/common-reviewer (reviewer)	Workspace ready	Prepared isolated checkout (merge_ref) at cfa11c9.
2026-07-17T20:25:28.435Z	ironloop/common-reviewer (reviewer)	Result captured	Changes requested; 1 blocking finding.
2026-07-17T20:25:28.435Z	ironloop/common-reviewer (reviewer)	Completed	Review completed and terminal status was persisted.
2026-07-17T20:29:53.628Z	ironloop/common-reviewer (reviewer)	Superseded	A newer PR head replaced this review (a6061d9).

Available commands

• @ironloopai help
• @ironloopai agents
• @ironloopai review
• @ironloopai review --agent <agent>

Run metadata

Admission: webhook accepted the request and IronLoop persisted reviewer state before this projection.

[coderabbitai]

📝 Walkthrough

Summary by CodeRabbit

• Tests
  • Added safeguards to detect unintended additions, removals, and duplicate exported LocalDev* types.
  • Improved checks for in-memory store type definitions, including conditional compilation scenarios.
  • Added coverage ensuring type detection ignores comments and string contents while recognizing supported visibility and trait modifiers.
• Refactor
  • Consolidated shared type-definition scanning and duplicate-detection utilities for more consistent architectural checks.

Walkthrough

Adds shared Rust scanner utilities for exported type definitions, migrates the InMemoryStore ratchet to them, and introduces a LocalDev type allowlist ratchet with duplicate and cfg-gating tests.

Changes

Anti-slippage ratchet scanning

Layer / File(s)	Summary
Shared scanner implementation crates/ironclaw_architecture/tests/ratchet_support/mod.rs	Adds occurrence tracking, workspace discovery, recursive Rust-source collection, cfg-aware duplicate detection, visibility/type scanning, and comment/string stripping.
InMemoryStore ratchet migration crates/ironclaw_architecture/tests/reborn_inmemory_store_ratchet.rs	Migrates inventory and self-tests to the shared scanner contracts and removes duplicated local helpers.
LocalDev type allowlist ratchet crates/ironclaw_architecture/tests/reborn_localdev_typename_ratchet.rs	Adds a frozen LocalDev* inventory, only-shrinks validation, scanner self-tests, same-file duplicate checks, and cfg-gated alias coverage.

Estimated code review effort: 4 (Complex) | ~45 minutes

Sequence Diagram(s)

sequenceDiagram
  participant RatchetTest
  participant collect_type_defs
  participant scan_type_defs
  participant duplicate_definitions
  RatchetTest->>collect_type_defs: scan crates for matching exported types
  collect_type_defs->>scan_type_defs: parse Rust source
  scan_type_defs-->>collect_type_defs: identifiers and cfg_gated flags
  collect_type_defs-->>RatchetTest: occurrence map
  RatchetTest->>duplicate_definitions: validate multiplicity
  duplicate_definitions-->>RatchetTest: duplicate definitions

Loading

Possibly related PRs

• nearai/ironclaw#6204: Refactors the InMemoryStore ratchet to consume the shared scanner utilities introduced here.

🚥 Pre-merge checks | ✅ 3 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Description check	⚠️ Warning	The PR body is detailed, but it does not follow the required template and omits major sections like Change Type, Linked Issue, Validation, and the review checklists.	Reformat the body to the template and fill the missing sections: Summary bullets, Change Type, Linked Issue, Validation, Security/DB/Blast Radius/Rollback, and review checklists.

✅ Passed checks (3 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title is Conventional Commits-style and clearly names the LocalDev ratchet freeze.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

---

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

• X
• Mastodon
• Reddit
• LinkedIn

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

[ironloopai]

❌ IronLoop Review: reviewer

Review at a glance

Verdict	Blocking	Notes	Inline	Head
❌ Changes requested	1	0	1	7a2d4e513642

Head: 7a2d4e513642817c457726e35e954fc9228ef504
Next: Fix the blocking findings, push the PR branch, then re-run this reviewer.

Run details

Status: Current
Needs human: no
Needs validation: no

Summary

The focused stack-layer review found one blocking correctness issue: the new architecture ratchet does not reliably distinguish Rust type definitions from non-code text and misses valid trait syntax.

Findings

Blocking: 1 / Notes: 0

Blocking findings

1. ❌ [MEDIUM] The line scanner does not enforce an exact Rust type inventory

Location: crates/ironclaw_architecture/tests/reborn_localdev_typename_ratchet.rs:157-186
This scans isolated lines without tracking comments or string literals. A line beginning pub struct LocalDevFoo inside a block comment or multiline/raw string is therefore counted as a definition; it can cause false CI failures or keep an allowlisted name present after the real type is deleted. Conversely, the declared trait coverage misses valid definitions such as pub unsafe trait LocalDevFoo. Because exact set membership is the sole behavior introduced here, use Rust-aware lexing/parsing and extend the self-test with block-comment, raw-string, and qualified-trait cases.

Developer follow-up

After fixing this feedback:

1. Push the fix to this PR branch.
2. Re-run this reviewer with @ironloopai review --agent reviewer if you only changed this reviewer's findings.
3. Re-run all reviewers with @ironloopai review when the fix may affect multiple areas.

[ironloopai]

This line-oriented scan is not an exact type-definition scanner: it counts pub struct LocalDevFoo inside block comments or multiline/raw strings, while missing valid declarations such as pub unsafe trait LocalDevFoo. That can either create false CI failures or let non-code text keep an allowlisted name alive after the real type is deleted. Please use Rust-aware lexing/parsing and add self-test cases for those contexts.

[ilblackdragon]

Addressed in ee85b4c by consolidating both §10 ratchets onto one hardened scanner (tests/ratchet_support/mod.rs) instead of duplicating a weaker copy: comments and plain/raw string literals are stripped by a minimal lexer before matching (so definition-shaped text in them neither adds a false positive nor keeps an allowlisted name alive), pub unsafe trait / pub(crate) unsafe trait / auto trait declarations are matched, and the scan is occurrence-preserving with a multiplicity check. That last part surfaced a real subtlety your comment predicted: composition's factory.rs defines the same LocalDev* alias twice under mutually exclusive #[cfg] gates (incl. rustfmt-split multi-line gates) — the duplicate check now exempts a name only when every occurrence is cfg-gated, and a mixed gated/ungated pair still fails. Self-tests cover comment/string fixtures, unsafe-trait forms, same-file duplicates, and both cfg-pair shapes.

[railway-app]

🚅 Deployed to the ironclaw-pr-6205 environment in ironclaw-ci-preview

Service	Status	Web	Updated (UTC)
ironclaw	✅ Success (View Logs)	Web	Jul 17, 2026 at 11:10 pm

[ilblackdragon]

✅ Ready for merge

Reviewed, all findings addressed, CI fully green (18 pass / 0 fail) on head ee85b4ca8. Stacked on #6204 — GitHub will retarget to main when that merges; the diff is only the ratchet slice.

What was done:

• Rebased onto test(reborn): freeze the InMemory*Store allowlist ratchet (§10) #6204's current head (the hardened InMemory-ratchet scanner).
• Review verdict: the §4.4/§10 frozen-inventory design matches the merged-sibling pattern; the frozen 31 verified against the stronger scanner (all live in src/, none dropped by the production-scope walk).
• IronLoop finding + my review finding, fixed in ee85b4ca8: instead of duplicating a weaker line-scanner, both §10 ratchets now share one hardened implementation (tests/ratchet_support/mod.rs): comment/string stripping (the reported false-positive/false-negative source), pub unsafe trait/auto trait matching, restricted visibility, occurrence-preserving scans, and multiplicity checking. That surfaced a real subtlety: composition factory.rs defines the same LocalDev* alias twice under mutually exclusive #[cfg] gates (including rustfmt-split multi-line gates) — the duplicate check exempts a name only when every occurrence is cfg-gated; a mixed gated/ungated pair still fails. All pinned by self-tests (comment/string fixtures, unsafe-trait forms, same-file duplicates, single- and multi-line cfg pairs, mixed pair).
• Local gates: both ratchet binaries (4+4 tests) + full architecture suite green; clippy -D warnings clean; pre-commit safety within budget. PR body updated for the expanded (shared-support) scope.

🤖 Generated with Claude Code

[coderabbitai]

Actionable comments posted: 2

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@crates/ironclaw_architecture/tests/ratchet_support/mod.rs`:
- Around line 16-19: Update the duplicate-detection logic around the
matched-definition cfg handling to prove that cfg predicates are mutually
exclusive before suppressing a duplicate; do not treat mere presence of #[cfg]
as sufficient, and account for definitions in separate modules. Preserve valid
mutually exclusive alias pairs, while keeping identical or overlapping
predicates—including the LocalDev fixture’s disabled-feature combination—as
duplicates. Add regression cases covering identical and overlapping gates, and
revise the documentation near the matched-definition model to describe the
behavior actually enforced.

In `@crates/ironclaw_architecture/tests/reborn_localdev_typename_ratchet.rs`:
- Line 48: Update the KEYWORDS constant in the LocalDev typename scanner to
include "union " so named Rust unions are inventoried alongside structs, enums,
traits, and type aliases. Add a regression fixture covering a pub union
LocalDev* declaration and assert it is detected, preserving the documented
§4.4/§10 freeze guarantees.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: ASSERTIVE

Plan: Pro Plus

Run ID: b03b519a-4e07-45aa-9269-4876aaf853b4

📥 Commits

Reviewing files that changed from the base of the PR and between d51cc14 and 1269253.

📒 Files selected for processing (3)

• crates/ironclaw_architecture/tests/ratchet_support/mod.rs
• crates/ironclaw_architecture/tests/reborn_inmemory_store_ratchet.rs
• crates/ironclaw_architecture/tests/reborn_localdev_typename_ratchet.rs

[coderabbitai]

🎯 Functional Correctness | 🟠 Major | 🏗️ Heavy lift

Prove cfg exclusivity before suppressing duplicates.

Line 45 equates “every occurrence has a #[cfg]” with “mutually exclusive.” Identical or overlapping gates—and unrelated gates in separate modules—are therefore hidden from both ratchets.

The current LocalDev fixture at Lines 230-236 also overlaps when inmemory-turn-state, libsql, and postgres are disabled. Preserve predicates and prove disjointness, or narrowly exempt explicitly reviewed alias pairs. Add identical/overlapping-gate regression cases that remain duplicates.

As per coding guidelines, “Comments and documentation that promise guarantees must match the behavior implemented in code and covered by tests.”

Also applies to: 34-46

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@crates/ironclaw_architecture/tests/ratchet_support/mod.rs` around lines 16 -
19, Update the duplicate-detection logic around the matched-definition cfg
handling to prove that cfg predicates are mutually exclusive before suppressing
a duplicate; do not treat mere presence of #[cfg] as sufficient, and account for
definitions in separate modules. Preserve valid mutually exclusive alias pairs,
while keeping identical or overlapping predicates—including the LocalDev
fixture’s disabled-feature combination—as duplicates. Add regression cases
covering identical and overlapping gates, and revise the documentation near the
matched-definition model to describe the behavior actually enforced.

Source: Coding guidelines

[coderabbitai]

🎯 Functional Correctness | 🟠 Major | ⚡ Quick win

Include Rust unions in the LocalDev* inventory.

pub union LocalDevState { ... } defines a named type but bypasses this scanner, violating the §4.4/§10 freeze. Add "union " and a scanner regression fixture.

Proposed fix

-const KEYWORDS: &[&str] = &["struct ", "enum ", "trait ", "type "];
+const KEYWORDS: &[&str] = &["struct ", "enum ", "trait ", "type ", "union "];

As per coding guidelines, every bug fix requires regression coverage and documented guarantees must match behavior.

Also applies to: 140-182

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@crates/ironclaw_architecture/tests/reborn_localdev_typename_ratchet.rs` at
line 48, Update the KEYWORDS constant in the LocalDev typename scanner to
include "union " so named Rust unions are inventoried alongside structs, enums,
traits, and type aliases. Add a regression fixture covering a pub union
LocalDev* declaration and assert it is detected, preserving the documented
§4.4/§10 freeze guarantees.

Source: Coding guidelines

[github-actions]

Coverage ratchet

Ratchet mode: ENFORCING

RATCHET PASS: global
  observed: 85.94% (304812 / 354688 lines)
  floor:    85.3% (tolerance 0.5pp -> effective floor 84.8%)
  denominator: 354688 lines now vs 320188 at floor capture (+34500 lines, +10.77%) — material change (>5%)

⚠️ 2 Reborn crate(s) have 0 int-tier coverage (target: 0) — ironclaw_prompt_envelope, ironclaw_scripts

Reborn integration-tier coverage

Line coverage (Reborn crates): 85.94% — 304812 / 354688 lines

Per-crate breakdown (62 crates, lowest-covered first)

Crate	Line %	Covered / Total
ironclaw_prompt_envelope	0%	0 / 88
ironclaw_scripts	0%	0 / 345
ironclaw_runtime_policy	31.75%	80 / 252
ironclaw_event_projections	43.31%	673 / 1554
ironclaw_observability	61.54%	16 / 26
ironclaw_authorization	62.46%	604 / 967
ironclaw_mcp	64.89%	595 / 917
ironclaw_triggers	65.44%	2142 / 3273
ironclaw_dispatcher	67.15%	92 / 137
ironclaw_filesystem	67.69%	3932 / 5809
ironclaw_memory	69.2%	773 / 1117
ironclaw_reborn_migration	71.57%	1551 / 2167
ironclaw_trust	72.88%	661 / 907
ironclaw_reborn_cli	74.19%	7010 / 9449
ironclaw_capabilities	74.36%	1685 / 2266
ironclaw_wasm_limiter	74.6%	47 / 63
ironclaw_reborn_event_store	74.67%	958 / 1283
ironclaw_extractors	74.72%	538 / 720
ironclaw_projects	76.48%	400 / 523
ironclaw_llm	78.25%	20193 / 25805
ironclaw_product_context	78.57%	11 / 14
ironclaw_process_sandbox	80.65%	671 / 832
ironclaw_wasm_product_adapters	80.71%	1448 / 1794
ironclaw_first_party_extensions	81.06%	5965 / 7359
ironclaw_memory_native	81.22%	3205 / 3946
ironclaw_events	81.43%	1539 / 1890
ironclaw_secrets	82.79%	2794 / 3375
ironclaw_reborn_identity	83.59%	433 / 518
ironclaw_processes	83.76%	939 / 1121
ironclaw_run_state	83.96%	424 / 505
ironclaw_wasm	83.97%	1011 / 1204
ironclaw_reborn_config	84.06%	1814 / 2158
ironclaw_auth	84.81%	3233 / 3812
ironclaw_turns	85.04%	13722 / 16136
ironclaw_host_api	85.34%	2695 / 3158
ironclaw_product_workflow	85.79%	10917 / 12725
ironclaw_network	86.12%	670 / 778
ironclaw_common	86.66%	1741 / 2009
ironclaw_slack_v2_adapter	86.79%	1806 / 2081
ironclaw_threads	86.93%	4708 / 5416
ironclaw_product_adapters	87.18%	3265 / 3745
ironclaw_skills	87.6%	4471 / 5104
ironclaw_hooks	87.78%	9921 / 11302
ironclaw_product_adapter_registry	88.06%	531 / 603
ironclaw_reborn_traces	88.19%	11946 / 13546
ironclaw_webui	88.42%	7333 / 8293
ironclaw_host_runtime	88.65%	17840 / 20123
ironclaw_reborn_composition	88.97%	75489 / 84850
ironclaw_extensions	89.38%	2971 / 3324
ironclaw_runner	89.5%	16989 / 18983
ironclaw_reborn_openai_compat	89.5%	3778 / 4221
ironclaw_approvals	90.18%	1598 / 1772
ironclaw_conversations	90.39%	3123 / 3455
ironclaw_event_streams	90.82%	1009 / 1111
ironclaw_resources	91.73%	4490 / 4895
ironclaw_loop_host	92.25%	15051 / 16316
ironclaw_attachments	93.06%	630 / 677
ironclaw_telegram_v2_adapter	93.91%	2592 / 2760
ironclaw_agent_loop	94.88%	9184 / 9680
ironclaw_safety	95.04%	3677 / 3869
ironclaw_outbound	95.59%	3556 / 3720
ironclaw_first_party_extension_ports	95.62%	3672 / 3840

This table itself is informational and never gates the PR on its own — not the percentage, not the per-crate holes, not the 0-coverage callout. A separate coverage ratchet (dry-run until enforce=true; see tests/integration/coverage-floor.toml) can fail the build on specific configured floors.

Exemptions (3 entry/entries excluded from the accounting above)

Module / Crate	Reason	Issue
crate: ironclaw_embeddings	v1-only: consumed only by root ironclaw (src/app.rs, src/tools/builtin/memory.rs, src/workspace/mod.rs, src/config/{mod,embeddings}.rs); no crates/* dependents. Covered by "Tests (Legacy)".	#5657
crate: ironclaw_gateway	v1-only: consumed only by root ironclaw (src/channels/web/platform/static_files.rs, src/channels/web/handlers/frontend.rs); no crates/* dependents. Covered by "Tests (Legacy)".	#5657
crate: ironclaw_tui	v1-only: consumed only by root ironclaw (src/main.rs, src/channels/tui.rs); no crates/* dependents. Crate's own doc comment confirms it bridges INTO v1, not Reborn. Covered by "Tests (Legacy)".	#5657

[ilblackdragon]

Post-rebase confirmation: after #6204's squash-merge, this branch was rebased onto main (only its two commits remain; conflict resolved) and the full main-targeted CI matrix is green — 56 pass / 0 fail on head 126925360. Still ready for merge.