# NousResearch/hermes-agent · skills_guard false positives and verdict handling fixes

## Open now

- 1 issue, 2 PRs open on GitHub.
- #37036: skills_guard false-positive: mksglu/context-mode blocked by 12 false DANGEROUS findings
- PR #15437: fix(tools): make skills_guard 'ask' verdict return actionable findings instead of generic block (#13686)
- PR #37040: fix(skills): cap documented-example findings one step instead of demoting all markdown (#37036)

Multiple issues and fixes address the skills_guard tool producing false-positive CRITICAL/HIGH findings on benign skill content, and the 'ask' verdict being incorrectly treated as a hard block rather than a confirmation prompt. Fixes also cover honoring skill ignore files and improving actionability of returned findings.

## TL;DR

- 8 member(s): 1 open issue(s), 2 closed issue(s), 2 open PR(s), 1 merged PR(s), 2 PR(s) closed without merging.
- A fix was merged to main in PR #97060. Verify it addresses your case.
- **Recommended:** read PR #97060 (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: `partially_resolved` · deciding PR(s): `pr-97060`. 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evidence (`STRUCTURAL_PARTIAL`): some members are connected by structural evidence (shared changed files, or a closing/superseded reference); the rest are joined by an automated relevance judgement over their text.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 `0.22`: 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

- #37036: skills_guard false-positive: mksglu/context-mode blocked by 12 false DANGEROUS findings

## Closed / duplicate issues

- #13686: skills_guard: agent-created skills hard-blocked by false-positive regex patterns — 'ask' verdict treated as block
- #60709: skills_guard: several rules produce false-positive CRITICAL/HIGH on benign skills (hard-blocks community installs)

## Prior & competing PRs

- **#97060: fix(skills-guard): benign skills no longer blocked by five false-positive scanner patterns (#60709, salvage #60750)**. merged to main · fix tier `verified`
  - maintainer direction: category `merged`; implication `superseded`
  - files: `tests/tools/test_skills_guard.py`, `tools/skills_guard.py`
- **#15437: fix(tools): make skills_guard 'ask' verdict return actionable findings instead of generic block (#13686)**. open
  - maintainer direction: category `uncategorized`; implication `needs_llm`
  - files: `tests/tools/test_skill_manager_tool.py`, `tests/tools/test_skills_guard.py`, `tools/skill_manager_tool.py`
- **#37040: fix(skills): cap documented-example findings one step instead of demoting all markdown (#37036)**. open
  - maintainer direction: category `wontfix_bydesign`; implication `wontfix`; refs `d188a47e9a`, `a4f1d4e876`, `d188a47e9a`, `#111334`
  - files: `tests/tools/test_skills_guard.py`, `tools/skills_guard.py`
- **#60750: fix(skills-guard): reduce false-positive CRITICAL/HIGH on benign skill patterns**. closed without merge
  - maintainer direction: category `merged_via`; implication `superseded`; refs `#97060`
- **#124428: fix(skills-guard): stop flagging template files, loopback/example hosts, and markdown table rows (#37036)**. closed without merge
  - maintainer direction: category `implemented_elsewhere`; implication `superseded`; refs `6e6d94b2e2`

## Fix status

- Merged to main in PR #97060 (cited above). No release data is tracked. Verify downstream.
- Fix last verified (merged) `2026-08-28`.
- Subsystem: `skills`