Phase 22: capability truth table + MCP scope freeze + release gates - #54
Merged
Conversation
Phase 22 closeout. Adds structure/CAPABILITY_TRUTH_TABLE.md with the canonical five-column status table (Capability | Status | Code Location | Tests | Mirror In cli-jaw) covering browser runtime cleanup/doctor, ChatGPT/Gemini/ Grok resolvers, action-intent / target-resolver (incl. send.click), answerArtifact, sourceAudit, frozen MCP browser tool scope, external CDP deferral, benchmark trajectory, and named release gates. Links the table from structure/INDEX.md. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Locks the agbrowse MCP browser-tool surface to the two implemented tools (browser_snapshot, browser_click_ref) and tracks the eight planned tools in NOT_IMPLEMENTED_BROWSER_TOOLS. Adds a strict JSON-Schema validator (validateBrowserToolInput) wired into mcp-server.callMcpTool so any unknown property, wrong type, missing required field, out-of-range number, bad enum, or non-matching ref pattern is rejected before provider work happens. Adds test/unit/phase22-mcp-freeze.test.mjs as a regression gate that fails if a new browser tool is registered without: - being added to BROWSER_TOOLS or NOT_IMPLEMENTED_BROWSER_TOOLS, AND - being mentioned in structure/CAPABILITY_TRUTH_TABLE.md. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Locks External / Remote CDP as deferred (experimental). Documents the contract that must exist before the deferral is lifted (auth, lifecycle ownership, policy boundary, trace-evidence parity, disconnect/origin spoofing tests) and the rules for any experimental code paths (EXPERIMENTAL marker, opt-in flag, truth-table sync, no README readiness claim). Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
…checks Add scripts/release-gates.mjs implementing 5 named gates: - typecheck: node --check on entry points + structure/check-doc-drift.sh (agbrowse is .mjs; no tsc) - tests: npm test - truth-table-fresh: 7-day mtime window + required-term presence - mcp-scope-frozen: enforces FROZEN_BROWSER_TOOL_NAMES via schema import - no-experimental-in-readme-ready-section: scans README ready/feature sections for forbidden experimental terms Wired via 6 package.json scripts (gate:typecheck, gate:tests, gate:truth-table-fresh, gate:mcp-scope-frozen, gate:no-experimental-in-readme-ready-section, gate:all). Gates are the named contract referenced by Phase 22 release claims; violating any one blocks a release. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Add npm run gate:all step to: - .github/workflows/release.yml (after fixture evals, before pack/publish) - scripts/release.sh (before pack/publish dry-run) - scripts/release-preview.sh (before pack dry-run) Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
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.
Rebased Phase 22 work onto current main (P02-P51 strict-checkjs already merged).
Commits (5)
Verification
Branch hygiene
After this merges, only main + chore/strict-migration remain (chore/strict-migration tracks future strict-migration work).
Co-authored-by: Copilot 223556219+Copilot@users.noreply.github.com