Investigate and harden Roam detectors, CLI/MCP result contracts, and evidence consumers when dogfooding or correcting incomplete, misleading, or inconsistent analysis. Use for a requested hardening pass, a suspected evidence defect, or unresolved Roam result/delivery state, not routine feature edits, product copy, or release approval.
Installs into .claude/skills of the current project.
Are you the author of Roam Evidence Hardening?
Add the live security badge to your README. It updates with every re-scan.
[](https://www.skillsdirectory.com/skills/gabrielmoreira-roam-evidence-hardening-agent-skills-mirror)
---
name: roam-evidence-hardening
description: "Investigate and harden Roam detectors, CLI/MCP result contracts, and evidence consumers when dogfooding or correcting incomplete, misleading, or inconsistent analysis. Use for a requested hardening pass, a suspected evidence defect, or unresolved Roam result/delivery state, not routine feature edits, product copy, or release approval."
---
# Shared project skill
The canonical body lives at `.agents/skills/roam-evidence-hardening/SKILL.md` in this repo (the cross-agent
location Codex reads natively). **Read that file now and apply it**; resolve its relative
references against that directory. Do not maintain a second body here — this stub exists
only because Claude Code does not read `.agents/skills`.
