rea-agents 6.9.0 advertised `^[^\uD800-\uDFFF]*$` on every local path
input. RE2 has no `\u` escape and Rust's `regex`/`jsonschema` crates refuse
non-scalar code points, so strict MCP clients and gateways (DeepSeek, RE2
gateways, Open Design) rejected the complete tools request and no REA tool
could run.
Move the rule into one runtime check, `rejectUnpairedSurrogates`, which
advertises nothing and still raises the `invalid_format` regex issue, so
CLI and MCP corrections keep reason, expected value, and message. The
objdump disassembly symbol already enforced the same rule with its own
refinement; it now shares the owner and the correction shape.
Dropping the pattern also brings inspect_android_jni_bindings within the
compact profile's per-schema budget, so compact delivery now keeps its
DEX dimension pattern. The descriptor admission test follows the actual
presentation instead of assuming compact always omits it.
Fixes#10238Fixes#10348Fixes#10198
Co-Authored-By: gensokyo-dev <gensokyo-dev@users.noreply.github.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* test: assert absolute-path facts without frozen sentences
Sibling path tests now require the field path, invalid_format, and regex
expectation, and they record that each negative case ran. NUL rejection
follows the argument that contains the byte instead of searching for the
word NUL.
* test: retain actionable NUL path diagnostics
---------
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
* feat(artifacts): select extraction entries and output destination
Refs morluto/rea#3490
* fix(artifacts): omit absent selection from execution parameters
* fix(artifacts): protect source directories and verify public selection
* fix(artifacts): keep optional selection inputs JSON-safe
* fix(artifacts): separate reader ownership from selected destinations
Validate complete reader inventory independently from collisions among selected output entries. Use occurrence-parent ownership instead of filename prefixes so independently owned PyInstaller entries remain selectable while nested-reader members remain refused. Validate advertised MCP schemas through the shared acceptance helper and add real PyInstaller/ASAR and excluded-entry regressions.
* fix(artifacts): advertise absolute extraction destinations
Reuse the shared host-absolute local path schema instead of an opaque refinement so SDK-converted JSON Schema preserves the runtime constraint. Extend the advertised-path and MCP-handler regression matrix for extract_artifact.output_root.
* fix(contracts): advertise unique extraction occurrence IDs
---------
Co-authored-by: gensokyo-dev <gensokyo-dev@users.noreply.github.com>
* feat(native): read PE and ELF export tables
Add a host-independent export-table reader on the shared native
container layer. PE images report each used export-address-table slot
with its ordinal, one row per name, its address or forwarder text, and
the export directory identity; the PE image reader now exposes its data
directories. ELF images report defined default/protected global, weak
and unique .dynsym symbols with type, binding and .gnu.version_d
version (hidden versions kept distinct, unreadable version tables
"unknown"). Unreadable entries are counted in issues and make coverage
partial; an ELF without section headers is unsupported.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* feat(native): inspect native export tables through CLI and MCP
Expose the reader as inspect_native_exports / inspect-native-exports:
one stable, digest-bound read of an explicit file with a caller-selected
input budget, typed failures for malformed, unsupported, oversized and
cancelled reads, and observed Evidence bound to the file path and
SHA-256. Mach-O inputs point to inspect_macho.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* docs(native): describe portable PE and ELF export inventories
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fix(native): bound export inventories and read header-backed PE tables
Address review of the export-table reader:
- Charge every retained row and its name, forwarder and version text
against shared 262144-row and 16 MiB allowances before storing it, so
repeated long names or PE aliases cannot exhaust memory.
- Map PE RVAs through the headers as well as sections; the PE image
reader now exposes its SizeOfHeaders mapping.
- Keep ELF exports with version "unknown" when .gnu.version lies outside
the file, and report an empty version-definition name as unresolved.
- Reject a .dynsym without its mandatory null entry as malformed.
- Decode the ELF section directory once.
- Correct the docs on which gaps drop rows and which keep them.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
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.
Reuse the schema-visible host path owner across session, firmware, managed, browser, Electron and Evidence-file inputs. Keep existing runtime and CLI path semantics.
Refs #5493.
