Codex delivery under spec access-control-prover. `semantic/obligations/paths.py` builds entry-to-function path records from the cached caller index; `mapping.py` names Stapler web methods and Jetty handlers as entries; `check_obligations` gains the opt-in premise `assume_entries_protected` with the label `unguarded_path`; `hunt/prover.py` proves a change with no model call and states its coverage; `benchmarks/valid/prover_arm.py` scores it with the valid-benchmark metrics. Two fixes after the first smoke run: a product file git holds empty is read, not unread (four changes failed on `__init__.py` and empty assets), and absent-input reasons are counted by name. Smoke on the 23 fix changes of the training pool: 7 pushes caught (30 percent, Wilson 16 to 51), 28 of 93 locations, precision 7 of 8 flagged, no instrument failure; three changes over the 120 second bound. Twelve of the 17 Jenkins fixes reach the fixed method but check no operation because the method is the operation; the spec now carries handler-named operations for delivery B. Co-Authored-By: Claude Fable 5.1 <[email protected]>
7.6 KiB
Implementation Plan
-
1. Paths from the caller index
-
1.1 Add
PathRecordandcompose_pathsinsrc/openultrasast/semantic/obligations/paths.py(P)- Upward walk through a
callers(path, function)callable, depth and fan-out bounds, stop at the first entry, chains kept per file, deduplicated; unreached targets returned - Done when
tests/test_obligation_paths.pycovers depth 1 to 3, unreached, fan-out cut and same-name files, andcheck_obligationsaccepts the records aspaths(one test runs the checker with them) - Requirements: 3.1, 3.2, 3.3, 3.4, 3.5, 3.6
- Boundary: obligations.paths
- Upward walk through a
-
1.2 Checker opt-in
assume_entries_protectedand labelunguarded_pathinsemantic/obligations/check.py- Default off;
_findingreceives the entry records' access by handler; evidence as the design states;findings_to_staticexplains the label;LABELSupdated - Done when the existing checker tests pass unchanged and a new test shows: with paths, no policy, no siblings,
the default gives no finding and the opt-in gives one
unguarded_pathfinding for a non-public entry and none for an entry whose record access ispublic - Requirements: 4.1a
- Boundary: obligations.check (opt-in only)
- Default off;
-
2. Entry recognisers
-
2.1 Stapler and Jetty recognisers in
mapping.py(P)- Public
do[A-Z]...methods on exposed types or with Stapler-typed or@QueryParameterparameters; JettyAbstractHandler,HttpServlet, JAX-RS resources; evidence strings;publiconly forUnprotectedRootAction - Done when
tests/test_mapping_frameworks.pynames the entries of a fixture Descriptor, rejects a private helper nameddoX, and the existing mapping tests still pass - Requirements: 1.1, 1.6, 1.7, 1.8
- Boundary: mapping
- Public
-
2.2 Symfony, Livewire and Laravel controller, Remix export and Tornado recognisers in
mapping.py(P)- Done when fixture files of each shape yield
routeentries with evidence and the negative fixtures yield none - Requirements: 1.2, 1.3, 1.4, 1.5, 1.7, 1.8
- Boundary: mapping
- Done when fixture files of each shape yield
-
3. Enforcement witnesses and facts
-
3.1
semantic/obligations/enforcement.pyproducing router and decoratorDischargewitnesses (P)- Laravel route middleware and controller
middleware/authorizeResource; SymfonyIsGranted,Security, security.yamlaccess_control; Tornadoauthenticated; names listed in the facts headers - Done when
tests/test_obligation_enforcement.pyshows a guarded Laravel route, a Symfony attribute and a Tornado decorator each produce one witness thatOrderDominanceaccepts for the handler's operation - Requirements: 2.1, 2.2, 2.5, 2.6
- Boundary: obligations.enforcement
- Laravel route middleware and controller
-
3.2 Operation and discharger rows for the pool's shapes in
ruleset/obligations/{java,php,python,typescript}.toml(P)- Java: credential lookups and outbound connection tests in
doTestConnection-style handlers, item reads and writes,checkPermission/checkAnyPermissionas path guards,hasPermissionas ownership check; Remix: session requirements as path guards, record reads by request id as protected reads needingownership_check; Tornado: handleropen/getas entries needingpath_guard; Jetty service start as privileged action. Every name checked in the clones or documentation, said in the header - Done when
tests/test_obligation_facts.pyloads the files and one operation and one discharger per language resolve - Requirements: 2.3, 2.4, 2.7
- Boundary: ruleset/obligations
- Java: credential lookups and outbound connection tests in
-
3.3 Handler-named operations:
handlersandframeworkfields in the facts loader,find_operations(..., handlers_of=), Stapler form-validation rows injava.toml, Tornado handler rows inpython.toml, Remix action rows intypescript.toml- Done when
tests/test_obligation_facts.pyloads a handler-named row,tests/test_obligation_operations.py(or the existing operations test file) yields the operation at the declaration line only for listed handlers, and the prover smoke on the 23 fixes reaches and checks the Jenkins form-validation methods - Requirements: 2.8
- Boundary: obligations.facts, obligations.operations
- Done when
-
4. The prover
-
4.1
hunt/prover.py:prove_changeover the hunterContext,Coverage, findings as hunt defects- Touched functions from head lines, build output left out, targets include direct callees holding operations, IRs of hop files passed to the checker, extra witnesses from enforcement, coverage counts by reason, no model
- Done when
tests/test_hunt_prover.pyshows: an unguarded Stapler handler callinglookupCredentialsgives one finding with entry, hops and missingpath_guard; acheckPermissionabove the call gives none; a guarded Laravel route gives none; an unrecognised framework gives coverage only; the finding dict haspath,line,family,function,evidence,id - Requirements: 4.1, 4.2, 4.3, 4.4, 4.5, 4.6, 6.2, 7.1, 8.1, 8.2
- Depends: 1.1, 2.1, 3.1, 3.2
-
5. The arm
-
5.1
benchmarks/valid/prover_arm.py: run over a pool slice with checkpoints, summarise with the valid metrics- Record envelope copied from
hunter_arm; instrument fields; resume;summary.jsonwith recall (pushes and locations, Wilson), false alarms per 100, precision, instrument failures, seconds per change and the list over 120 s, per-repository and per-framework coverage,uncoveredrepositories; refuse on zero-byte reads - Done when
tests/test_prover_arm.pyruns a two-change fixture pool end to end and the summary carries every field above; a record with zero bytes read makessummarizerefuse - Requirements: 5.1, 5.2, 5.3, 5.4, 5.5, 6.1, 7.2
- Depends: 4.1
- Record envelope copied from
-
6. Measurement (maintainer, not Codex)
-
6.1 Run the arm on
benchmarks/valid/train/pool-train.tomlover the training clones; writebenchmarks/measurements/<date>-access-control-prover/with NOTES, summary, coverage table; compare with the DSPy and hunter arms of the same pool- Requirements: 5.2, 5.3, 7.2
Implementation Notes
- Delivery A (2026-10-14) implements paths, the opt-in checker premise, Stapler/Jetty/Servlet/JAX-RS entries, the Java facts, the deterministic prover and the checkpointed arm. No commit was made.
- Task 3.2 stays open for the PHP, Python and TypeScript facts. Task 4.1's delivery A core is delivered;
its full-spec checkbox stays open for delivery B's enforcement witnesses and guarded Laravel acceptance
case. This delivery intentionally passes
extra_witnesses=()as required by the maintainer's brief. - Default checker behavior is preserved. Only opt-in checking includes a caller's statement witness, using the existing dominance contract at the outgoing call site. The prover checks each complete path independently so a guarded chain cannot hide an unguarded chain to the same operation.
- Java additions were checked against actual Git HEAD source objects in the sparse training clones; checked call sites are named in the Java TOML header. No network or model call was made.
- Mapping accepts cached source texts and an opt-out from external manifest scanning. Default callers retain their previous behavior; the delivery A prover uses source entries and its Context caches.
- Delivery A review was performed in the main context because the brief forbids model calls. The exact requested static gate passed (format, Ruff and mypy); the exact pytest gate passed with 112 passed, zero failed, exit 0. UV_NO_SYNC=1 was set; UV_CACHE_DIR was placed under /tmp because the default cache is outside the writable sandbox. The unchanged unjustified-path checker test remains intact.