docs: map hooks to OWASP LLM Top 10 2026 - #38
Merged
Conversation
Adds docs/owasp-llm-top-10-2026.md: a per-risk mapping of every hook against the official 2026 list, with honest verdicts (N/A named as N/A), a 2025 to 2026 what-moved section, and a pointer to the Agentic Top 10. Also links the doc from the README and fixes the stale hooks docs URL.
Rebased on main after both hooks merged: LLM01 gains config-guard (the agent cannot rewrite hook config) and instructions-audit (poisoned instruction files lock the session); LLM08 names instructions-audit for the inbound half instead of listing it as a wanted contribution.
karanb192
force-pushed
the
docs/owasp-2026-mapping
branch
from
August 18, 2026 03:36
9385349 to
f27a5a7
Compare
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Adds
docs/owasp-llm-top-10-2026.md: a per-risk mapping of every hook in this repo against the OWASP Top 10 for LLM Applications, 2026 edition. The doc carries the 10-row verdict table, a 2025 to 2026 what-moved section, and a scoping pointer to the OWASP Top 10 for Agentic Applications with the four ASI risks these hooks land on. The README gets a short OWASP mapping section linking the doc, and the stale Claude Code hooks docs URL now points to https://code.claude.com/docs/en/hooks.Why
OWASP shipped the 2026 edition in early August: the official resource page announces it on August 3, 2026, and the OWASP project page says published August 4, 2026. Canonical entry files: GenAI-Security-Project/GenAI-LLM-Top10. Excessive Agency moved up to LLM03 in this edition, and that is the risk this repo exists for, so the repo should say precisely where it lands on the new list.
Mapping philosophy: honest verdicts. Risks a runtime tool-call hook genuinely covers are marked covered; risks it cannot touch (training-time poisoning, model supply chain, misinformation) are marked N/A instead of stretched into a claim; gaps that fit the hook model (rate limiting, output scanning, dependency validation) are named as wanted contributions.
Tests
npm teston this branch (docs-only change, suite untouched):