waterflane · waterflane · Sep 8, 2026 · Sep 8, 2026
@@ -6,7 +6,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/).
 Feature releases follow [Semantic Versioning](https://semver.org/spec/v2.0.0.html),
 and Python distribution versions follow PEP 440.
 
-## [Unreleased]
+## [0.5.1] - 2026-09-05
 
 ### Added
 
@@ -17,36 +17,6 @@ and Python distribution versions follow PEP 440.
 - Added opt-in Bridge 2.0 tracked `build`/`update` jobs, correlated
   `$/progress` notifications, cooperative cancellation, schema capabilities,
   and a normative Bridge 2 JSON Schema. Bridge 1.0/1.1 remain supported.
-
-### Changed
-
-- Classified authentication, authorization, missing credential, quota, rate
-  limit, model, configuration, timeout, and service failures so terminal
-  provider-wide failures are not retried per file.
-- Made semantic and repository-map analyzer identity depend on provider/model
-  and analysis contracts rather than an OpenAI-compatible endpoint. Legacy
-  `+base.<sha256>` records migrate on update without model calls.
-
-### Fixed
-
-- Enforce the caller's expected snapshot inside the index workflow, revalidate
-  it immediately before atomic publication, and check cancellation at the
-  manifest activation boundary.
-- Return Bridge index timeouts at the caller's deadline while retaining
-  cooperative cleanup and its writer lock as tracked background work.
-- Bound Bridge progress delivery with one writer task, coalesce synchronous
-  cumulative progress bursts, detect sustained client backpressure, drain every
-  completed semantic task, and preserve typed provider causes through semantic
-  and map aggregation.
-- Preserve the prior active generation on cancellation, failure limits,
-  provider circuit opening, source drift, or progress backpressure.
-- Return safe typed Bridge index errors without provider bodies, credentialed
-  URLs, absolute paths, tracebacks, or exception representations.
-
-## [0.5.1] - 2026-09-05
-
-### Added
-
 - Added verified Tree-sitter declaration extraction for JavaScript, TypeScript,
   Java, C#, Go, Rust, C, C++, PHP, and Ruby, including language-specific
   ownership, signatures, modifiers, exports, and parser diagnostics.
@@ -60,6 +30,12 @@ and Python distribution versions follow PEP 440.
 
 ### Changed
 
+- Classified authentication, authorization, missing credential, quota, rate
+  limit, model, configuration, timeout, and service failures so terminal
+  provider-wide failures are not retried per file.
+- Made semantic and repository-map analyzer identity depend on provider/model
+  and analysis contracts rather than an OpenAI-compatible endpoint. Legacy
+  `+base.<sha256>` records migrate on update without model calls.
 - Prioritized exact current-source identifiers, verified declarations, inferred
   regions, and explicit task facets during discovery; verified symbol questions
   now prefer compact declaration and dependency ranges.
@@ -69,24 +45,6 @@ and Python distribution versions follow PEP 440.
 - Expanded semantic planning so every required symbol receives response budget,
   oversized work splits before dispatch, and validated chunks can resume after
   interruption.
-
-### Added
-
-- Added immutable, model-free repository discovery application contracts for
-  deterministic candidate preparation, read-only expansion, verified source
-  reads, and canonical context packaging.
-- Added the generic ContextForge bridge protocol v1 ADR, normative JSON Schema,
-  persistent workspace-bound JSON-RPC 2.0 service over bounded UTF-8 NDJSON
-  stdio, and a runnable generic client example.
-- Added explicit protocol negotiation and compatibility failures for missing or
-  unsupported v1 handshakes, cooperative JSON-RPC request cancellation, typed
-  source-identity drift, deterministic preparation/selection identities, and
-  bounded concurrent requests with serialized responses.
-- Added bridge architecture, CLI, troubleshooting, security, stream,
-  cancellation, read-only, repository-truth, and source-verification guidance.
-
-### Changed
-
 - Preserved exact identifier ranges through discovery review and required
   sufficient multi-symbol context before accepting model finalization.
 - Preserved shadowed builtin dependencies, TypeScript destructuring defaults,
@@ -99,6 +57,24 @@ and Python distribution versions follow PEP 440.
 - Corrected polyglot declaration ownership, callable classification, source
   ranges, and coverage reporting across supported grammar variants.
 
+### Fixed
+
+- Enforce the caller's expected snapshot inside the index workflow, revalidate
+  it immediately before atomic publication, and check cancellation at the
+  manifest activation boundary.
+- Return Bridge index timeouts at the caller's deadline while retaining
+  cooperative cleanup and its writer lock as tracked background work.
+- Bound Bridge progress delivery with one writer task, coalesce synchronous
+  cumulative progress bursts, detect sustained client backpressure, drain every
+  completed semantic task, and preserve typed provider causes through semantic
+  and map aggregation.
+- Preserve the prior active generation on cancellation, failure limits,
+  provider circuit opening, source drift, or progress backpressure.
+- Return safe typed Bridge index errors without provider bodies, credentialed
+  URLs, absolute paths, tracebacks, or exception representations.
+- Updated installed wheel and source-distribution smoke validation for Bridge
+  2.0 while retaining explicit checks for 1.0 and 1.1 compatibility.
+
 ## [0.5.0] - 2026-08-24
 
 ### Added

@@ -28,11 +28,10 @@ commands.
 </p>
 
 > [!IMPORTANT]
-> ContextForge is pre-alpha software. The unreleased API adds bounded failure
-> policies, JSONL progress, and opt-in Bridge 2 tracked index jobs while the
-> package version remains `0.5.1`. Discovery benchmarking is experimental and
-> its results should be reviewed alongside the recorded provider, model,
-> configuration, and source snapshot.
+> ContextForge is pre-alpha software. Version `0.5.1` adds bounded failure
+> policies, JSONL progress, and opt-in Bridge 2 tracked index jobs. Discovery
+> benchmarking is experimental and its results should be reviewed alongside the
+> recorded provider, model, configuration, and source snapshot.
 
 ## Why ContextForge
 

@@ -68,7 +68,7 @@ This roadmap describes broad milestones only. It does not promise dates.
   read-only guarantees inside ContextForge while consumers own model selection.
 - [x] Keep MCP independent and preserve its existing read-only protocol.
 
-### v0.5.1 repository intelligence maintenance release (release prepared)
+### v0.5.1 repository intelligence and host-integration release (complete)
 
 - [x] Add verified Tree-sitter declaration extraction for ten common polyglot
   languages and publish explicit relationship-coverage limitations.
@@ -81,8 +81,6 @@ This roadmap describes broad milestones only. It does not promise dates.
 - [x] Adapt structured model requests to truncation and reported runtime context
   limits without weakening strict validation.
 
-### Unreleased host-integration hardening (complete)
-
 - [x] Add independent fail-fast and bounded-failure index policies while
   preserving the existing `--fail-on-error` contract.
 - [x] Classify provider failures, stop retrying terminal authentication, quota,

@@ -1,8 +1,9 @@
-# Bridge API consumer follow-ups
+# ContextForge 0.5.1 Bridge API consumer follow-ups
 
-The retrieval and Bridge 1.1 changes in ContextForge do not rewrite downstream
-capsule assembly. Consumers such as `dsh-contextforge` should track the following
-separately when migrating from Bridge 1.0:
+The retrieval, Bridge 1.1, and opt-in Bridge 2 changes shipped together in
+ContextForge 0.5.1. They do not rewrite downstream capsule assembly. Consumers
+such as `dsh-contextforge` should track the following when migrating from Bridge
+1.0:
 
 - omit empty `compressed_summary` values instead of serializing placeholders;
 - remove alphabetical fallback candidates when discovery reports
@@ -11,6 +12,12 @@ separately when migrating from Bridge 1.0:
   Markdown table or source excerpt;
 - use Bridge 1.1 match evidence and registered expansion candidate IDs rather than
   rebuilding candidate lists client-side;
+- negotiate Bridge 2 only when ContextForge should own index build/update,
+  writer-lock, staging, cancellation, and atomic publication lifecycle;
+- consume correlated progress continuously and treat its sequence as monotonic
+  rather than contiguous because cumulative burst snapshots may be coalesced;
+- after `REQUEST_TIMEOUT`, allow tracked background cleanup to release the index
+  lock before starting another mutation;
 - keep raw prompts, provider diagnostics, workspace permissions, approval policy,
   and other Harness metadata out of user-visible answers and packaged context.
 

@@ -107,5 +107,6 @@ diagnostics, nested-ignore, and CLI usability improvements; it does not claim
 future GUI, remote transport, orchestration, or source-mutation functionality.
 Version 0.5.0 adds the trusted-local, model-free, read-only bridge v1 without
 changing the existing model-assisted discovery or MCP semantics. Version 0.5.1
-adds verified polyglot declarations, resumable semantic coverage, bridge
-protocol 1.1, and stricter exact-symbol and dependency-aware discovery.
+adds verified polyglot declarations, resumable semantic coverage, stricter
+exact-symbol and dependency-aware discovery, bounded failure policies, JSONL
+progress, provider circuit breaking, and opt-in Bridge 2 tracked index jobs.
@@ -2,7 +2,7 @@
 
 ## Status
 
-Accepted for the unreleased ContextForge API following 0.5.1.
+Accepted for ContextForge 0.5.1.
 
 ## Context
 
@@ -77,4 +77,4 @@ reserved for unexpected defects.
 - Bridge 2 has narrowly scoped index and provider authority; it still cannot
   write source, mutate Git, or execute arbitrary commands.
 - Persisted index, manifest, and record schemas remain version 2; progress
-  remains version 3 and the package version remains 0.5.1 until release work.
+  remains version 3 in ContextForge 0.5.1.