# NousResearch/hermes-agent · ZWJ emoji falsely flagged as prompt injection threat

## Open now

- 1 issue, 4 PRs open on GitHub.
- #59492: False positive: ZWJ emoji (U+200D) in context files (SOUL.md/AGENTS.md) is flagged as injection and drops the whole file
- PR #59503: fix(security): skip ZWJ in legitimate emoji sequences during injection scan
- PR #59622: fix(agent): surface threat-scanner block warnings to users
- PR #59652: fix(prompt-builder): surface threat-scanner context blocks to the user
- PR #59668: fix(agent): preserve ZWJ (U+200D) in context files (#59492)

A threat/injection scanner incorrectly flags Zero Width Joiner (U+200D) characters in context files, causing legitimate emoji sequences and files like SOUL.md to be dropped. Multiple fixes attempt to exempt ZWJ within valid emoji grapheme clusters from the invisible-unicode detection rule, with at least one related change to surface scanner block warnings to users.

## TL;DR

- 12 member(s): 1 open issue(s), 1 closed issue(s), 4 open PR(s), 0 merged PR(s), 6 PR(s) closed without merging.
- No fix has been merged to main; one or more PRs were attempted (see below).
- **Recommended:** review the prior/competing PR(s) below. A fix was attempted but not merged; your case may extend or revive that work.

## Maintainer resolution verdict

- Internal-tracker verdict over this cluster's issue members: `unresolved`. This is distinct from the fix status above. Weigh it yourself.

## How this cluster was grouped

- Link evidence (`STRUCTURAL_PARTIAL`): some members are connected by structural evidence (shared changed files, or a closing/superseded reference); the rest are joined by an automated relevance judgement over their text.
- Members outside that structural core may belong to a different underlying problem.
- Pairwise link density `0.16`: the share of member pairs that carry a link, counting a structural link in full and a relevance-only link at 0.36. `1.00` means every pair is linked.

## Open related issues

- #59492: False positive: ZWJ emoji (U+200D) in context files (SOUL.md/AGENTS.md) is flagged as injection and drops the whole file

## Closed / duplicate issues

- #18581: 🤸‍♀️ SOUL.md blocked by ZWJ emoji — cartwheel gymnast triggers prompt injection filter

## Prior & competing PRs

- **#35808: fix: allow ZWJ (U+200D) in legitimate emoji sequences**. closed without merge · fix tier `suggested`
  - maintainer direction: category `uncategorized`; implication `needs_llm`
  - _suggested-tier fix detail withheld; re-request with `?tier=all` to include it._
- **#12673: fix: allow ZWJ inside emoji grapheme clusters in context/memory/skills scanners**. closed without merge
  - maintainer direction: category `uncategorized`; implication `needs_llm`
- **#18644: fix: remove ZWJ from invisible char blocklist to allow compound emoji in context files**. closed without merge
  - maintainer direction: category `duplicate_of`; implication `duplicate`; refs `#18616`
- **#59503: fix(security): skip ZWJ in legitimate emoji sequences during injection scan**. open
  - maintainer direction: category `uncategorized`; implication `needs_llm`
  - files: `tests/tools/test_cronjob_tools.py`, `tools/cronjob_tools.py`, `tools/threat_patterns.py`
- **#59622: fix(agent): surface threat-scanner block warnings to users**. open
  - maintainer direction: category `uncategorized`; implication `needs_llm`
  - files: `agent/prompt_builder.py`, `tests/agent/test_prompt_builder.py`
- **#59652: fix(prompt-builder): surface threat-scanner context blocks to the user**. open
  - maintainer direction: category `duplicate_of`; implication `duplicate`; refs `#59622`, `#59612`, `#59625`, `#59622`, `#59622`
  - files: `agent/prompt_builder.py`, `tests/agent/test_prompt_builder.py`, `tests/tools/test_threat_patterns.py`, `tools/threat_patterns.py`
- **#59668: fix(agent): preserve ZWJ (U+200D) in context files (#59492)**. open
  - maintainer direction: category `duplicate_of`; implication `duplicate`; refs `#59503`, `#59492`, `#59503`, `#12673`
  - files: `tests/agent/test_context_file_emoji.py`, `tools/threat_patterns.py`
- **#59701: fix(agent): allow ZWJ emoji sequences in threat scanner (#59492)**. closed without merge
  - maintainer direction: category `duplicate_of`; implication `duplicate`; refs `#59503`, `#59503`, `#12673`, `#59492`
- **#59710: fix(threat_patterns): stop flagging legitimate ZWJ emoji as invisible-unicode injection**. closed without merge
  - maintainer direction: category `duplicate_of`; implication `duplicate`; refs `#59503`, `#59492`, `#59503`, `#12673`
- **#59925: fix(threat-scanner): exempt ZWJ emoji sequences from invisible unicode flag (#59492)**. closed without merge
  - maintainer direction: category `duplicate_of`; implication `duplicate`; refs `#59503`, `#59492`, `#59701`, `#59668`, `#59710`, `#59923`

## Fix status

- No merged fix on record.
- Subsystem: `agent`