Commit Graph
11 Commits
Author SHA1 Message Date
bdf0be87c6 fix(mcp): name response-byte and startup ceilings in recovery (#6005)
* fix(mcp): name response-byte and startup ceilings in recovery

Name REA_MCP_MAX_RESPONSE_BYTES, its effective value, the 10485760 byte
floor, and process restart when the stdio response budget is hit, and
state that the setting does not raise Node's string ceiling. Name fixed
MCP startup phase budgets when those deadlines fire, and keep Codex's
30 second startup figure as client guidance.

Co-authored-by: morluto <morluto@users.noreply.github.com>

* fix(mcp): distinguish string ceilings and keep dossier recovery

Name REA's reserved string limit separately from Node's ceiling, restore
detail:summary discovery for an oversized complete evidence bundle, and
apply the doctor deadline sentence when binary_session itself times out.

Co-authored-by: morluto <morluto@users.noreply.github.com>

* fix(mcp): keep retained SDK errors inside the text budget

Quoted SDK validation diagnostics expand when recovery JSON is escaped
into MCP text. Shorten only that preview so the retained-evidence
reference still arrives as parseable error content.

Co-authored-by: morluto <morluto@users.noreply.github.com>

* fix(mcp): keep export advice when a summary exceeds the response budget

An oversized get_evidence_bundle summary no longer repeats detail: summary.
export_evidence_bundle still writes the retained bundle outside one MCP frame,
including when a larger byte budget cannot raise the string ceiling.

Co-authored-by: morluto <morluto@users.noreply.github.com>

* style(mcp): format the SDK diagnostic preview return

oxfmt rejected the fitted-preview ternary in EvidenceMcpServer.ts, which failed Static checks and the CI required gate.

Co-authored-by: morluto <morluto@users.noreply.github.com>

---------

Co-authored-by: Cursor Agent <cursoragent@cursor.com>
Co-authored-by: morluto <morluto@users.noreply.github.com>
2026-10-22 05:48:56 +08:00
gensokyo-dev aa51e43439 perf(mcp)!: default to compact input schema advertisements (#5975)
Use compact MCP input advertisements by default while preserving explicit full, canonical validation, output contracts and complete Evidence. Require an explicit selected catalog for runtime identity and honor both profiles in CLI status.

Real ripgrep/Zellij SDK tasks, independent ELF oracle, actual client trials, profile parity workflows and CI are recorded in experiments/analysis-efficiency/mcp-input-catalog.md. SDK token estimates improve; actual Codex token savings and general latency/memory gains are not established.

Refs #5861

BREAKING CHANGE: MCP input advertisements default to compact. Set REA_MCP_INPUT_SCHEMA_PROFILE=full for complete advertised guidance and nested constraints.
2026-10-22 00:17:11 +08:00
gensokyo-devandKaoru0822 c23e161629 perf(mcp): record catalog identities for all schema profiles (#5415)
Share compiled schema conversion while recording all four built-in catalog profiles. Preserve canonical hashes, output validation and custom-budget fallback. Refs #723.

Co-authored-by: Kaoru0822 <208281322+kaoru0822-kitauji@users.noreply.github.com>
2026-10-20 10:09:28 +00:00
morluto c5fb86d194 fix(mcp): bind catalog identity to output schema profiles
Include the selected output advertisement profile while retaining canonical output validation in the identity. Reuse the SDK output projection and select doctor expectations from both profiles. Correct the compact input error documentation.
2026-10-15 01:41:31 +08:00
187f113f79 fix(identity)!: preserve invocation uncertainty and advertised catalogs
Capture the current entrypoint once without inventing a path. Preserve unknown registration ownership through setup and doctor, refuse unsafe registration plans, and retain exact pinned-launcher readiness checks. Hash the same full or compact schema projection advertised by each MCP server and use that identity in doctor comparisons.

BREAKING CHANGE: server.command_path may be null when the invocation path is unavailable. Registration and setup statuses, and doctor classifications, may report unknown.

Reviewed-PR: #3230, #3015
Co-authored-by: qorvr <330186599+qorvr@users.noreply.github.com>
Co-authored-by: Hemang Varshney <218273197+hemv-857@users.noreply.github.com>
2026-10-14 16:10:46 +08:00
morlutoand1052326311 ae433f0983 fix(mcp): bound retained doctor stderr while draining output
Retain a 64 KiB diagnostic prefix while counting and draining the complete stderr stream. Report truncation and avoid manufacturing a replacement character at a truncated UTF-8 boundary.

Validation: all nine doctor tests pass, including real noisy child, malformed bytes, canonical catalog and startup deadline controls.

Based on #3121.

Co-authored-by: 1052326311 <65798732+1052326311@users.noreply.github.com>
2026-10-14 13:58:17 +08:00
morluto 80de3145e4 fix(mcp): align advertised contracts and retain workflow evidence
Use one named input-schema projection for clients and catalog identity, preserve dialect declarations, and bound compact schema depth. Keep canonical SDK output validation when output advertisements are disabled. Release request-owned state on cancellation and carry upstream limitations through composed workflows. Preserve doctor full-output envelopes and validate navigation facts at their owner.

Reviewed fixes: #2088, #2108, #2513, #2684, #2693, #2695, #2836, #2864, #2868, #2877, #2933, #2998, #3002, #3015.

Validation: focused schema, SDK output-validation, context, and workflow tests passed; no schema dialect or canonical output validation was relaxed.
2026-10-14 08:32:00 +08:00
Ysopwnable ea92328775 fix(mcp): validate doctor probe against its discovered output contract (#1461)
Validate the target-free probe against the named definition discovered from the complete production catalog. Preserve initialized server and inventory observations when the SDK rejects the probe, and retain the schema failure reason, absolute deadline and owned-child cleanup.

The negative stdio replay regression fails on original main and passes the fix; actual production-boundary fault injection confirms rejection. Nine doctor/help tests, serial full static checks and real small-target CLI/MCP parity pass. Successful doctor reports remain byte-identical. No performance or model-token reduction is claimed.

Closes #1454. Refs #1065, #723, #1433.
2026-10-10 14:33:57 +08:00
morluto f72931d9dd test: eliminate fixture races and prune redundant scaffolding (#849)
* test: eliminate fixture races and prune redundant scaffolding

* fix(test): publish launcher capture readiness without renaming
2026-10-07 12:25:37 +08:00
morluto b8a9da0c0f refactor(mcp): make analysis states unrepresentable (#468)
* refactor(config): parse capability settings into exact states

* refactor(runtime): encode replay and process outcomes exactly

* refactor: make analysis states unrepresentable

* refactor(mcp): register canonical schemas with SDK

* fix: preserve capability and catalog invariants

* fix(build): remove stale compiled modules

* fix(replay): update real verifier policy dependency

* fix(package): verify the complete MCP catalog

* fix(test): supply complete replay reload policy

* refactor(contracts): centralize output schema lookup

* refactor(mcp): delegate protocol boundaries to SDK

* fix(browser): approve real source capture

* fix(mcp): preserve catalog guidance
2026-08-10 02:28:11 +08:00
morluto 513d9b1555 test: overhaul deterministic suite architecture 2026-07-29 22:00:38 +00:00