# NousResearch/hermes-agent · SOUL.md injection phrase triggers false-positive warnings

## Open now

- Nothing open on GitHub.

These items address a threat scanner or prompt builder incorrectly flagging SOUL.md as a security risk because it documents a canonical injection phrase rather than using one maliciously. The work involves adjusting handling so documented phrases load with a warning or are allowed through without being blocked.

## TL;DR

- 3 member(s): 0 open issue(s), 1 closed issue(s), 0 open PR(s), 1 merged PR(s), 1 PR(s) closed without merging.
- A fix was merged to main in PR #113173. Verify it addresses your case.
- **Recommended:** read PR #113173 (and any competing PRs below) before filing. Confirm the merged change covers your symptom.

## Maintainer resolution verdict

- Internal-tracker verdict over this cluster's issue members: `resolved` · deciding PR(s): `pr-113173`. This is distinct from the fix status above. Weigh it yourself.

## How this cluster was grouped

- Link evidence (`STRUCTURAL_SPANNING`): every member is connected by structural evidence: shared changed files, or a closing/superseded reference.
- Pairwise link density `0.67`: 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.

## Closed / duplicate issues

- #112570: \[bug\] Context-file threat scanner false-positives on SOUL.md that documents the canonical injection phrase

## Prior & competing PRs

- **#113173: fix(prompt_builder): SOUL.md that documents an injection phrase loads with a warning instead of being blocked (#112570, supersedes #112572)**. merged to main · fix tier `verified`
  - maintainer direction: category `merged`
  - files: `agent/context_file_sources.py`, `agent/prompt_builder.py`, `tests/agent/test_context_file_sources.py`, `tests/agent/test_prompt_builder.py`, `website/docs/developer-guide/prompt-assembly.md`, `website/docs/guides/use-soul-with-hermes.md`, `website/docs/user-guide/features/context-files.md`, `website/docs/user-guide/security.md`
- **#112572: fix(security): allow documented context injection phrases**. closed without merge
  - maintainer direction: category `superseded_by_pr`; implication `superseded`; refs `#113173`, `#112572`

## Fix status

- Merged to main in PR #113173 (cited above). No release data is tracked. Verify downstream.
- Fix last verified (merged) `2026-09-17`.