v1.7.0: harness reach (MCP tools + integrate), web hardening, cost credibility, CLI errors#70
Merged
Merged
Conversation
…r (v1.6.1) The per-call telemetry log passed structlog-style kwargs (model=, tokens=, ...) to the stdlib logger returned by get_logger(). Under the CLI, which enables root INFO via configure_logging(), this raised TypeError: Logger._log() got an unexpected keyword argument 'model' right after the model responded and tokens were billed, so every live `cloudwright design`/`modify` produced a traceback and no spec. Latent since v1.1.0: pytest leaves the root logger at WARNING, so info() short- circuits and the line never emitted or crashed under tests. Switch both the Anthropic and OpenAI providers to %-style stdlib formatting. Side effect: LLM telemetry (model, latency, tokens, cache hits) now actually emits. Regression: test_llm_telemetry_logging.py runs both providers under configure_logging() and asserts no crash.
…, spec schema_version
…viction, container auth guard, logging on launch
… OSCAL download, rebuilt bundle
… for 11 coding agents
…r-alternative + ratio confidence, coverage doc
…rness integrate command
…st confidence-tier docstring
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Implements the full July 2026 product audit fix list plus the harness-reach extension, in two waves of isolated-worktree work on top of the v1.6.1 hotfix.
Harness reach (the differentiators now reach every coding agent)
review,compliance/oscal,planare now MCP tools. 13 of 14 popular coding agents are MCP clients (Claude Code, Cursor, Cline, Windsurf, Copilot, Zed, Codex CLI, Junie, Kiro, Antigravity), so one server puts cloudwright's design-time checks in all their agent loops. MCP surface: 22 tools / 9 groups (was 18/6).cloudwright integrategenerates the exact MCP config per client (mcpServers/ VS Codeservers/ Zedcontext_servers/ Codex TOML), a CLI-pipe path for Aider (not an MCP client), and an AGENTS.md/CLAUDE.md/GEMINI.md rules block that gates infra through cloudwright. Grounded in sourced research (docs/integrations.md).Audit fixes
asyncio.to_thread), 1 MB body + component-count caps, rate-limiter bucket eviction, container auth guard (CLOUDWRIGHT_REQUIRE_AUTH),configure_logging()on all launch paths.prices_as_of(catalog vintage) split fromestimated_on, per-alternative confidence incompare, real regional price rows when present (else medium-confidence multiplier), a catalog-backed ratio.docs/provider-coverage.mdstates coverage plainly.--jsonmode.pip-auditgate +check_version_sync.pygate,RELEASING.md,SECURITY.md, specschema_version.Verification
/api/reviewreturns an offline critique with no key.Reviewed by Xavier Puspus