jonah/safe-ai-tools
1
0
Fork 0
Code Issues Pull requests 2 Projects Releases Packages Wiki Activity Actions

chore: land 0.2.0 on main (PR #2 merged into its stacked base) #3

Merged
jonah merged 2 commits from feat/core-and-redact-proxy into main 2026-10-05 21:23:23 +02:00
Conversation 1 Commits 2 Files changed 20 +583 -21
jonah commented 2026-10-05 21:19:33 +02:00
Owner
Copy link

PR #2 was stacked on feat/core-and-redact-proxy and merged there, so main is still 0.1.1.

This brings main up to 0.2.0: the accuracy harness, the synthetic test set, the 400 on non-text content parts, and the fix for the fail-open content gate. The diff against main is the single commit b21263c.

🤖 Generated with Claude Code

PR #2 was stacked on feat/core-and-redact-proxy and merged there, so main is still 0.1.1. This brings main up to 0.2.0: the accuracy harness, the synthetic test set, the 400 on non-text content parts, and the fix for the fail-open content gate. The diff against main is the single commit b21263c. 🤖 Generated with [Claude Code](https://claude.com/claude-code)
feat: accuracy harness + synthetic test set (stacked on #1) (#2)
Some checks failed
CI / policy (pull_request) Failing after 3s
Details
CI / python (pull_request) Successful in 21s
Details
CI / checks (pull_request) Successful in 1m16s
Details
b21263c7eb 
Stacked on #1: base is `feat/core-and-redact-proxy`. Merge #1 first, then retarget this to `main`.

## What
- Seeded NL/DE/EN test-set generator with exact gold spans, incl. checksum-failing look-alikes as unlabelled negatives.
- `python -m redact_proxy.eval`: span-level recall, precision, F1, leak rate per entity.
- Hermetic CI test: 99% recall floor on checksum entities.
- README accuracy table (spaCy md, 1,000 docs per language).
- **Behaviour change:** image/audio/file content parts return 400 by default (they were forwarded unredacted). `allow_non_text_parts: true` opts out.
- 0.2.0.

## Measured (synthetic, so an upper bound)
Leak rate: NL 16.5%, DE 7.4%, EN 8.8%. Weak: bare postcode/PLZ (33-47%), credit card (63-76%), phone (78-88%).

## Not done
`gliner2-pii` backend, `--verifier`, Gretel/ai4privacy corpora, lg-model comparison. Docker, Forgejo CI and live Ollama still unverified (see #1).

🤖 Generated with [Claude Code](https://claude.com/claude-code)

https://claude.ai/code/session_01JZCbAYmY516SAe7umoLKWt
Co-authored-by: CodeAndCanvas728 <cedar-elk-01@icloud.com>
Reviewed-on: #2
jonah commented 2026-10-05 21:19:35 +02:00
Author
Owner
Copy link

🤖 Auto-review status · jonah/safe-ai-tools PR #3

⏭️ Superseded — a newer push is being reviewed instead.

Last updated: 2026-10-05T19:27:41.459Z

<!-- opencode-status --> **🤖 Auto-review status** · `jonah/safe-ai-tools` PR #3 ⏭️ **Superseded** — a newer push is being reviewed instead. *Last updated: 2026-10-05T19:27:41.459Z*
Merge branch 'main' into feat/core-and-redact-proxy
Some checks failed
CI / checks (pull_request) Successful in 1m21s
Details
CI / policy (pull_request) Failing after 3s
Details
CI / python (pull_request) Successful in 23s
Details
69deb2a90f
jonah merged commit 4dcd2d7039 into main 2026-10-05 21:23:23 +02:00
jonah deleted branch feat/core-and-redact-proxy 2026-10-05 21:23:23 +02:00
Sign in to join this conversation.
Reviewers
No reviewers
Labels
Clear labels
No items
No labels
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
jonah/safe-ai-tools!3
No description provided.