default

scored

afi / afi · cli reviewed proxy at 9b51f95e with Default profile / OpenAI · gpt 5.6 terra high on afi 0.30.0

Measurements

Unavailable values retain their exported state and reason

Quality

Recall 35%

Model-judged precision N/A

F1 N/A

Tier 1 recall 33.3%

Tier 2 recall 14.3%

Tier 3 recall 50%

Tier 4 recall 50%

Security recall 0%

Defect recall 46.2%

Maintainability recall 0%

Performance recall 33.3%

Median anchor distance 0

Worst anchor distance 0

Refusals a judge overturned Not recorded

Review outcomes

Distinct matched defects 7

Defects missed 13

Findings 7

Unkeyed findings 0

Intended findings 0

Carried findings Not recorded

Cost, usage, and speed

Total bill $0.2338

Review bill $0.2338

Judge bill N/A

Tokens 1262

Duration 2m 11s

Execution and identity

Lifecycle, reviewer, build, settings, and normalized execution limits

Lifecycle: completed
Lifecycle reason: Not recorded
Started: 31 Aug 2026, 15:19
Finished: 31 Aug 2026, 15:21
Reviewer: afi
Reviewer tool: afi · cli
Configuration: default / b9ed88e87764
Configuration ID: config-778af814f6d5c0ac
Build: afi 0.30.0 / b0f313c0b59a66ecc7612396dc8db0ea5da13a7a
Comparison: comparison-5e375ce1a3d1f0fc
Build ID: build-33906dcd701a0660
Cohort ID: cohort-07363c4dd57b37b6
Harness: bench 1 / 631c4107752a97104719622800c9815e030613cf dirty
Adapter: afi 1 / sha256:a1a935298956020ee6d767a756847abb4c00694c88c2eb80d454886ebc4acf8c
Exit code: 0
Wall time: 131
Deadline exceeded: No
Budget: $8.00
Billing limit: $8.00

Usage

Normalized reviewer-tool usage projection

Input tokens: 12
Output tokens: 1250
Cached input: 54111
Reasoning tokens: 6698
Requests: 4
Models counted: Not recorded

Judge

Normalized assessment summary

Judge data is N/A for this run

Findings

Review text is escaped; normalized decisions follow each finding

Finding f1

bug

A blob committed by an in-flight fetch has not yet been linked into the index, so this unconditional “unreferenced” branch can delete it during the `commit`→`link` window; the fetch then links a… A blob committed by an in-flight fetch has not yet been linked into the index, so this unconditional “unreferenced” branch can delete it during the `commit`→`link` window; the fetch then links a missing blob and fails when it reopens it to serve the response.

bug

Category: bug Confidence: Not recorded

services/proxy/src/sweep.rs:119

Evidence quote

Not recorded

Suggested fix

Not recorded

matchedDefect 18 (locality)

Finding f2

bug

The sweep treats every file in `incoming` as abandoned, but active `BlobWriter`s (and temporary index writes) use this directory; a scheduled or manual sweep can unlink an active download and make… The sweep treats every file in `incoming` as abandoned, but active `BlobWriter`s (and temporary index writes) use this directory; a scheduled or manual sweep can unlink an active download and make its later rename/commit fail.

bug

Category: bug Confidence: Not recorded

services/proxy/src/sweep.rs:232

Evidence quote

Not recorded

Suggested fix

Not recorded

matchedDefect 17 (locality)

Finding f3

bug

forget` runs even for `dry_run=true`, deleting index entries despite the endpoint promising not to remove anything; the next real sweep will then treat the still-present blobs as unreferenced and… forget` runs even for `dry_run=true`, deleting index entries despite the endpoint promising not to remove anything; the next real sweep will then treat the still-present blobs as unreferenced and evict them.

bug

Category: bug Confidence: Not recorded

services/proxy/src/sweep.rs:140

Evidence quote

Not recorded

Suggested fix

Not recorded

matchedDefect 12 (locality)

Finding f4

bug

The on-demand route calls `Sweeper::sweep` directly instead of `run`, bypassing the `running` mutex; it can overlap the periodic sweep and both passes can make deletion decisions from stale totals.

bug

Category: bug Confidence: Not recorded

services/proxy/src/routes/admin.rs:76

Evidence quote

Not recorded

Suggested fix

Not recorded

matchedDefect 6 (locality)

Finding f5

bug

Sorting `Duration` ascending orders the youngest blobs first, not the oldest, so an over-capacity sweep evicts recently written eligible artifacts while retaining the oldest ones.

bug

Category: bug Confidence: Not recorded

services/proxy/src/sweep.rs:102

Evidence quote

Not recorded

Suggested fix

Not recorded

matchedDefect 1 (locality)

Finding f6

bug

values` returns reclaimed bytes before removed-blob count, whereas `COUNTERS` declares blob count before bytes, so both new Prometheus metrics publish each other’s values.

bug

Category: bug Confidence: Not recorded

services/proxy/src/metrics.rs:94

Evidence quote

Not recorded

Suggested fix

Not recorded

matchedDefect 13 (locality)

Finding f7

performance

The full synchronous filesystem walk and unlink pass runs directly on a Tokio worker thread; on a single-core deployment, a large or slow-mounted cache blocks that worker from serving requests for… The full synchronous filesystem walk and unlink pass runs directly on a Tokio worker thread; on a single-core deployment, a large or slow-mounted cache blocks that worker from serving requests for the duration of every sweep.

performance

Category: performance Confidence: Not recorded

services/proxy/src/sweep.rs:84

Evidence quote

Not recorded

Suggested fix

Not recorded

matchedDefect 11 (locality)

Source artifacts

Open a row to inspect or download evidence from this atomically published local run

findings present

Open verified source artifact

Open this path at the exported commit