You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@airflow.apache.org by "potiuk (via GitHub)" <gi...@apache.org> on 2026/05/04 21:37:13 UTC

[PR] feat(privacy-llm): refine PII contract + wire skill-side protocol [airflow-steward]

potiuk opened a new pull request, #50:
URL: https://github.com/apache/airflow-steward/pull/50

   ## Summary
   
   Two follow-up commits on top of the merged privacy-llm foundation (#48): the contract refinement (reporter NOT redacted; only non-collaborator third parties) and the skill-wiring layer (every Gmail-touching skill now references the canonical protocol).
   
   ### `8a82fbf` — refine PII contract
   
   The reporter sent the security@ mail and is operationally known to the security team (CVE credit, reply threads, sync comments). Their own identity flows through the agent's context as-is. **What gets redacted** is PII the reporter discloses about *other people* — third-party researchers, victims, named individuals other than the reporter — UNLESS the named individual is already a collaborator on the `<tracker>` repo (resolved via `gh api repos/<tracker>/collaborators`, the same source-of-truth as the prompt-injection rule), in which case their identity is already public/known and stays in the clear.
   
   Renamed the type code `R/reporter` → `N/name` everywhere (code + tests + docs) — the prefix `R` for "reporter" while the contract says reporters aren't redacted is a foot-gun.
   
   ### `d9fa976` — wire skill-side protocol
   
   - **`tools/privacy-llm/wiring.md`** (new) — single source-of-truth for the skill-side protocol: Step 0 pre-flight, redact-after-fetch (resolve collaborators → identify third-party PII → filter → `pii-redact`), reveal-before-send. Skill `SKILL.md` files link here rather than copying the protocol. Includes the table of skills that follow this pattern and explicit edge-case handling (collaborator-lookup fail, reporter-is-collaborator, lost mapping file, etc.).
   - **`tools/gmail/operations.md`** — privacy-LLM contract bolted onto every `get_thread` call (redact after fetch) and every draft-create call (reveal before send) so the protocol is unmissable for future skill authors.
   - **`projects/_template/privacy-llm.md`** — new *Redaction configuration* section with three adopter-tunable knobs:
     - `collaborator_source` (default `<tracker>`)
     - `collaborator_exemption` (default `enabled`; flip to `disabled` for stricter posture)
     - `redaction_field_types` (default all six)
   - **Skill files wired** with a Step 0 *Privacy-LLM contract* bullet pointing at `wiring.md`: `security-issue-import` (full pre-flight + redact + reveal), `-sync`, `-invalidate`, `-cve-allocate`, `-import-from-md`.
   
   ## Test plan
   
   - [x] `pytest` (redactor) — 48/48 pass
   - [x] `prek run --all-files` — clean
   - [ ] Visual review of `wiring.md` — the canonical pattern reads cleanly as the doc skill authors will land on.
   - [ ] Walk through `security-issue-import` Step 0 + Step 4 + Step 7.4 end-to-end — confirm the redact / reveal call points are at the right spots.
   - [ ] Confirm the `collaborator_exemption: disabled` knob description is clear enough that a PMC could use it for a stricter-posture deployment.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: commits-unsubscribe@airflow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] feat(privacy-llm): refine PII contract + wire skill-side protocol [airflow-steward]

Posted by "potiuk (via GitHub)" <gi...@apache.org>.
potiuk merged PR #50:
URL: https://github.com/apache/airflow-steward/pull/50


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: commits-unsubscribe@airflow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org