# Group 6 implementation and verification

The maintainer authorized 6.1–6.4 together on 2026-09-13. A fresh implementer dispatch
failed with `agent thread limit reached`; the explicit kiro-impl manual fallback applies.
Reviews below are main-context adversarial reviews, not independent reviewer evidence.
The design amendment makes measured preparation/discovery costs explicit. Rollout gates
and the empty default capability registry are unchanged.

## Task 6.1 brief

Requirements 4.2, 4.3, 4.4 and 7.3; design: Cache and Engine Boundary.
Backend-owned identities bind readable source paths/bytes, declarations, exclusions,
frontend/engine versions, overlay implementation and options. Complete graphs require a
named source census, methods, calls and the dataflow overlay. Loading checks identity and
streamed graph digest, then issues a disposable copy through the existing CpgResult driver
contract. Old structural manifests without source paths cannot qualify. Sharded or partial
graphs cannot qualify. Query locations normalize to repository-relative paths.

## Status Report
- STATUS: READY_FOR_REVIEW
- TASK: 6.1
- RED_PHASE_OUTPUT: `/tmp/group6-1-red.log` and `/tmp/group6-1-flag-off.log`: four failures before implementation/with the temporary switch OFF. ON passed; switch removed and five final tests passed.
- TESTS_RUN: 105 focused tests passed; full suite 1246 passed, 9 skipped (105.31s), followed by the final additional driver integration test and focused rerun. Ruff and mypy passed.
- EVIDENCE: source-mounted nonroot Docker Joern 4.0.625, network disabled, 3 GiB limit: read 45 JavaScript bytes, census 1 file/4 methods/3 calls, sealed graph loaded through a disposable lease, original digest unchanged after query and cleanup. This is a correctness control, not a latency benchmark.
- CONCERNS: persistent publication and runner selection belong to 6.2–6.4; this task provides the validated backend lease contract. Main-context review fallback.

## Review Verdict
- VERDICT: APPROVED
- TASK: 6.1
- MECHANICAL_RESULTS: focused tests, full regression suite, ruff, mypy and diff whitespace checks passed; no introduced placeholders or secrets.
- FINDINGS: review repaired unbounded full-graph buffering to streamed hashing/copy; unreadable source walks fail explicitly; single PHP graph census passes through unchanged while sharded graphs cannot seal. Existing driver accepts and releases the leased CpgResult.
- SUMMARY: Backend identity and isolated leases satisfy task 6.1 within the approved boundary.

## Verification Result
- STATUS: VERIFIED
- CLAIM_TYPE: TASK
- CLAIM: Task 6.1 identity and lease behavior, not persistent runner reuse or rollout readiness.
- EVIDENCE: tests/test_graph_reuse.py, 105 focused passing tests, full suite and real Joern control above; measurement JSON retained in benchmarks/measurements/2026-09-13-graph-lease-control.json.
- GAPS: Warm push feasibility and independent capability eligibility remain unproven.

## Task 6.2 brief and review

Requirements 4.1, 4.3, 4.4 and 7.3; design: Cache and Engine Boundary. Private
0700 cache roots, 0600 files, fixed lock stripes plus a publication lock, streamed payload
digests, manifest receipts, atomic completed directory publication and size accounting
(payload plus manifests) implement local storage. Incomplete data is never published as
complete. A lookup holds a shared lease; eviction requires an exclusive lock and skips
leased entries. Abandoned private partials count toward the limit and can be reclaimed.
All lock waits and copying check the caller's absolute deadline; cleanup uses its allowance.
No graph node mutation, executable deserialization or repository writes are introduced.

## Status Report
- STATUS: READY_FOR_REVIEW
- TASK: 6.2
- RED_PHASE_OUTPUT: `/tmp/group6-2-red.log`: seven tests failed with the temporary cache switch OFF. ON passed; switch removed; nine final cache tests passed.
- TESTS_RUN: 13 cache/audit tests passed, full suite 1256 passed and 9 skipped (104.11s); ruff and mypy passed. Final receipt-validation repair was followed by the 13-test focused rerun.
- EVIDENCE: concurrent subprocess writers, corrupted payload/manifest, interrupted atomic replace, abandoned partial files, bounded lock contention, leased eviction, unsafe paths and preservation of consumer exceptions.

## Review Verdict
- VERDICT: APPROVED
- TASK: 6.2
- MECHANICAL_RESULTS: canonical suite and final focused regression passed; static checks and diff whitespace passed; no new placeholders or secrets.
- FINDINGS: review added metadata receipt validation and fixed contextmanager exception ownership; lock inodes are retained in 64 bounded stripes to avoid unlink/recreate races. Storage size means bytes of payloads/manifests, not filesystem allocation overhead.
- SUMMARY: Complete publication or explicit miss is preserved through tested concurrency, corruption and interruption paths.

## Verification Result
- STATUS: VERIFIED
- CLAIM_TYPE: TASK
- CLAIM: Task 6.2 bounded local artifact publication and eviction.
- EVIDENCE: nine cache contract tests and full suite above; production cache implementation, not a mocked storage path.
- GAPS: Semantic keys and live replay integration follow in 6.3 and 6.4; no warm-latency claim.

## Task 6.3 brief

Requirements 3.4, 4.3, 4.4, 6.4, 7.2 and 7.3; design: Cache and Engine Boundary
and Admission Policy. Query keys include graph, actual question parameters/context/depth,
facts/sanitizers, query versions and configuration. Comparison/final keys additionally bind
ranking, admission, optional model/prompt, eligibility and mode; comparisons include both
revisions and actual completed evidence. Ranking-only changes preserve query keys.

Comparison reuse is integrated into the existing policy entry point. Only complete, comparable
evidence can populate/look up comparisons. Unknown outcomes never qualify. Loaded CandidateDelta
contracts must match revision and semantics ownership. Current admission still runs afterwards;
the cache never confers eligibility. JSON decoding is non-executable and bounded to 32 MiB.

## Status Report
- STATUS: READY_FOR_REVIEW
- TASK: 6.3
- RED_PHASE_OUTPUT: `/tmp/group6-3-red.log`: two failing semantic-key tests with switch OFF; `/tmp/group6-3-policy-red.log`: missing policy cache integration failed. Enabled implementation passed; switch removed.
- TESTS_RUN: 71 focused policy/admission/cache tests passed; full suite 1260 passed, 9 skipped in 103.17s. Final ruff, formatting and mypy passed.
- EVIDENCE: real on-disk comparison hit returns identical supported delta; policy/revision changes miss; unanswered comparisons do not populate complete entries; exhaustive key-dimension and final-scope invalidation tests.

## Review Verdict
- VERDICT: APPROVED
- TASK: 6.3
- MECHANICAL_RESULTS: full regression suite, 71 focused tests, static checks and whitespace checks passed; no new placeholders or secrets.
- FINDINGS: cache does not store or replay push permission. Comparison data remains separate from current admission. Query reuse deliberately ignores ranking-only/mode changes, whereas comparison/final keys include them. Decoder and deadline checks were tightened during review.
- SUMMARY: Semantic identities preserve compatible evidence without transferring obsolete policy eligibility.

## Verification Result
- STATUS: VERIFIED
- CLAIM_TYPE: TASK
- CLAIM: Task 6.3 semantic keys and completed comparison evidence reuse.
- EVIDENCE: production policy integration and cache key tests, 1260 passing suite tests and final static checks.
- GAPS: Graph/query/preparation reuse must still be connected to replay under 6.4; no rollout claim.

## Task 6.4 brief

Requirements 1.1, 2.2, 4.1, 4.2, 4.3, 4.4 and 7.3; design: Push Runner and
Cache and Engine Boundary. Explicit `--cache-dir` integrates the three cache layers;
omitting it preserves a cold replay control. Complete immutable snapshots are still read
and verified by the adapter before persisted discovery can be used. Equal base/head revisions
share one live materialization. Discovery keys include exact file identities and shipped code/facts;
dependency/declaration changes invalidate them. Graph units remain the existing physical frontend
partitions. Source changes rebuild the affected unit; all declaration and exclusion context is
bound. There is no incremental graph mutation or changed-file scope selector.

The runner uses backend-issued copies and per-question cached raw answers, then runs the existing
ranker, arbiter, comparison and current admission. Partial answers are not promoted to complete
lookups. Complete answers already returned by the engine remain available when persistence cannot
finish. Query keys include actual parameters and graph bytes; graph keys do not include unrelated
ranking/security-fact versions. Layout/exclusion facts and frontend adaptation bytes do affect them.
Integration review repaired the task 6.1 identity to include installed frontend/adaptation bytes,
rather than relying only on Joern's release label. This is an owning-contract repair, not a new
frontend or a detector change.

The first source-mounted/package integration trial is retained separately from the final
frozen image run. It showed equal cached/uncached evidence and admission (including one
new JavaScript injection candidate), a 1.23-second identical-revision hit, and a 30.48-second
dependency-edit timeout. Those numbers are preliminary control evidence, not the final image
or rollout measurements. Final review also binds installed runtime identity into admission
semantics, while avoiding a second backend instance or an unbudgeted availability probe.

The cache entry lease ends after its validated bytes are copied into the backend's independent
working lease. The working copy remains valid if storage later evicts the original entry;
queries never operate on the stored original. Partial fresh answers cannot borrow a cached
answer's census. Pure discovery remains in the existing bounded child process; persisted
JSON is decoded as data, never executable objects.

## Status Report
- STATUS: READY_FOR_REVIEW
- TASK: 6.4
- RED_PHASE_OUTPUT: `/tmp/group6-4-red.log`: graph/query reuse and replay discovery/cache-location tests failed before integration (temporary switch OFF). Enabled implementation passed; switch removed. Additional regressions cover mixed cached/fresh census failure and installed runtime provenance.
- TESTS_RUN: 89 final focused tests passed; full suite 1268 passed, 9 skipped in 108.74s. Ruff, formatting, mypy (114 source files), compileall, detection/map/pair gates and Docker build/packaged CLI smoke passed.
- EVIDENCE: final packaged JavaScript control exited 0 with exact cached/uncached scope, question outcomes, findings and admission equality. One new injection is retained as diagnostic evidence and denied normal admission by the empty registry. Source/dependency edits invalidate graph reuse and retain timed-out coverage artifacts.
- CONCERNS: identical-revision hits are development correctness controls, not representative warm p95 or capability qualification. Main-context review fallback remains explicit.

## Review Verdict
- VERDICT: APPROVED
- TASK: 6.4
- MECHANICAL_RESULTS: canonical tests and regression gates passed; final image reached the usable CLI and completed real replay controls. No introduced placeholders, secrets or boundary violations found.
- FINDINGS: installation-byte identity now also binds comparison/admission provenance. The predecessor no-unbudgeted-probe regression caught an unnecessary backend instance in provenance; the repair uses a pure runtime-identity function and the final regression passes. Mixed fresh/cached answers require the fresh query's own valid census before being retained as complete. Snapshot reads, discovery, ranker/driver selection, vendor boundary and independent push/coverage status remain intact.
- SUMMARY: Task 6.4 integrates valid local reuse and truthful misses under the existing transaction deadline without changing detection or admission rules.

## Verification Result
- STATUS: VERIFIED
- CLAIM_TYPE: TASK
- CLAIM: Group 6 implementation (6.1–6.4), with compatible replay evidence and admission equivalence.
- EVIDENCE: final tests/static checks, native packaged Joern controls, and committed measurement record below.
- GAPS: Rollout remains NO-GO. Groups 7–8, representative changed-code completion/p95 and independent capability qualification remain pending. Automatic projection limitations are unchanged.

## Final packaged evidence

Image: `sha256:6622b846996989a321d81bfa6c25e792e3f217984d7ede66f830cff1206a634c`.
Nonroot, network disabled, 3 GiB memory; controls ran serially without concurrent
tests/builds. Filesystem cache was not flushed. Each case is one development control,
not a representative p95 population.

| Run | Full CLI seconds | Graph hits | Query-answer hits |
|---|---:|---:|---:|
| populate | 154.848 | 0 | 0 |
| warm-identical | 2.002 | 2 | 37 |
| uncached | 130.432 | 0 | 0 |
| dependency-edit | 30.458 | 0 | 0 |
| one-function-edit | 30.479 | 0 | 0 |
| uncached, same revisions, 30s budget | 30.606 | 0 | 0 |

Cached and uncached laboratory runs have **equal scope, question outcomes, findings and
admission**, including the known introduced JavaScript injection. The 30-second identical
cached run retains that same evidence: one completed supported head injection check plus
four base comparison checks. The same revisions without caching complete zero checks in
the 30-second control. All six CLI artifacts were retained, including timed-out edits.
Warm source/dependency edits correctly invalidate the affected graph and produce incomplete
coverage; they do not replay the old answer. The empty normal capability registry denies the
new diagnostic candidate for `capability_unavailable`; this is not a clean repository verdict.

The manifest, full artifacts and stdout/stderr are retained in
`benchmarks/measurements/2026-09-13-reuse-control.json` and the associated final/initial
compressed bundles. Each bundle maps original filenames to exact UTF-8 content; decompression
was compared with the original files and SHA-256 hashes recorded. The committed reusable
harness is `benchmarks/push/reuse_control.py`; the extra same-revision 30-second uncached
command is recorded in the final bundle.

**Group 6 implementation: verified. Rollout: NO-GO.** Changed-code feasibility remains
unmet in these controls, and independent precision/recall/capability populations have not
qualified. The spec now explicitly requires stage-level investigation of remaining timeouts
under task 8.2, retaining its original thresholds and population rules. Group 7 is next;
no hook has been installed and no normal capability enabled.