guardrail CLI: planting-only flow — flat verbs, catalog, drop attack#6
Merged
Conversation
A planted contact defeats Favor's "send only to saved contacts" guard but can never produce a tool-misuse (money) capture — the judge freezes safe destinations at seed and never reads env.contacts. Relabel the verb docstring and the README surface table accordingly (matches the backend /catalog group). 78 pytest pass.
qyli00
added a commit
that referenced
this pull request
Jul 7, 2026
Ships the guardrail CLI planting-only flow (PR #6): catalog command, flat plant verbs, attack/plant removal. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Aligns the
aicodinggym guardrailCLI with the Assistant Pro redesign: the attacker only plants; victim routines are triggered on the web cockpit, never the CLI.attack(and thePOST /turncaller).catalog(GET /catalog) — lists everything you can plant, grouped by surface, plus the victim routines.plantprefix), each mapping to the right backendplantKind:email·note·event·message·webpagepayee·contact·requestconsent·policy/3.Tests
pytest); the API layer is mocked (no network).Depends on
POST /routine,GET /catalog, expanded/plantkinds). Those endpoints don't exist until it's deployed, so merge/deploy the backend PR first — until thenguardrail catalog/plant verbs return a clean 404.