docs: add rationale docs for CrewAI, AutoGen/AG2, Vercel AI SDK, Pydantic AI#6
Merged
Conversation
…dantic AI 30 rationale docs (one per shipped YAML pack) across four new SDK categories, covering all 47 new rules; the rulebook now documents 164 rules. Front-matter severity/confidence/scope match the shipped YAML; references are OWASP LLM Top 10 IDs, with CVE-grounded threat models and adversarial gap analysis. Regenerated POLICY_INDEX.md + four per-category indexes (check_rulebook --strict and gen_index --check both green). gen_index.py: register the four SDKs in SDK_ORDER/LABEL/FULL + the legend, and make the index write LF-safe and Python 3.9-compatible (write_text's newline= kwarg is 3.10+ and crashed regeneration under the repo's default python3).
Validate this PR's docs against the matching trustabl-rules PR before either merges: both jobs check out trustabl-rules at a branch matching this PR's head_ref when one exists, else main (push to main has an empty head_ref, so main always validates against the production pack).
trustabl
approved these changes
Jun 5, 2026
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
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Adds the paired rationale docs for the four new SDK rule packs — 30 docs (one per shipped YAML pack) under
docs/Policy/{crewai,autogen,vercel_ai,pydantic_ai}/, covering all 47 new rules. The rulebook now documents 164 rules total.Each doc's front-matter
severity/confidence/scopematches the shipped YAML exactly;referencesare OWASP LLM Top 10:2025 IDs. The code-execution / SSRF docs carry CVE-grounded threat models (CrewAI CVE-2026-2275/2287/2286/2285; Pydantic CVE-2026-46678/25580) plus the adversarial "what this policy does not cover" gap analysis (e.g. CrewAIBaseTool-subclass gap, Vercel.js/name-rule gaps, Pydantic RunContext-strip false-negative).Index + tooling
POLICY_INDEX.mdand the four new per-category indexes.check_rulebook.py --strictandgen_index.py --checkboth pass (0 warnings, 164 rules).tools/gen_index.py: registered the four SDKs inSDK_ORDER/SDK_LABEL/SDK_FULL+ the ID-prefix legend, and made the index write LF-safe and Python 3.9-compatible —write_text'snewline=kwarg is 3.10+ and was crashing regeneration under the repo's defaultpython3.Companion PRs
No Jira ticket supplied; add
Refs: TR-XXXif one applies.