Skip to content
Development
Skill

/cyber-evidence

Captures redacted, provenance-aware cybersecurity observations with shared status fields, independent ground truth, and honest limitations. Use when recording exercise evidence, findings, validation results, or security-review handoffs.

From plugin
agent-skills-standard
570200 skills21 agents33 commands1 MCP
Install
$ npx -y skills add hoangnguyen0403/agent-skills-standard --skill cyber-evidence --agent claude-code

How it fires

How this skill gets triggered: by you, by Claude, or both.

  • Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.Auto-invocation is when the right skill fires by itself at the right moment, driven by a FLOW.md router and a hook, instead of you invoking it by name. It is the difference between a skill being installed and a skill actually getting used.Read the full definition →
  • You can call itInvoke it directly when you want it.
  • Slash command/cyber-evidence

Context preview

The summary Claude sees to decide when to auto-load this skill.

Captures redacted, provenance-aware cybersecurity observations with shared status fields, independent ground truth, and honest limitations. Use when recording exercise evidence, findings, validation results, or security-review handoffs.

SKILL.md

cyber-evidence.SKILL.md
name: cyber-evidence
guardrail: true
description: Captures redacted, provenance-aware cybersecurity observations with shared status fields, independent ground truth, and honest limitations. Use when recording exercise evidence, findings, validation results, or security-review handoffs.
metadata:
  labels: [cybersecurity, evidence, provenance]
  triggers:
    keywords: [security evidence, evidence record, redacted evidence, finding status, ground truth, evidence gap]

Cyber Evidence

**Priority: P0 (CRITICAL)**

Record what was observed, how it was sourced, and what remains unknown; never turn absence of proof into proof.

Structure

cyber-evidence/
├── SKILL.md
├── references/evidence-record.md
└── evals/evals.json

Workflow

1. Bind record to `engagement_scope_ref`, `skill_version`, and `source`. 2. Capture `observed_at`, observer, method, and evidence references; redact secrets and sensitive payloads. 3. Set exactly one `finding_status`: `confirmed`, `suspected`, `blocked`, `not-tested`, or `false-positive`. 4. State limitations, missing runtime controls, synthetic-data boundaries, and accountable owner. 5. Preserve independently held ground truth; compare it only through an independent adjudicator. 6. Update the canonical `artifacts/security-review.md` when a workflow chain already owns that artifact. Shared record fields: `engagement_scope_ref`, `skill_version`, `source`, `observed_at`, `finding_status`, `evidence_refs`, `limitations`, `accountable_owner`.

Rules

  • Hashes and timestamps support provenance; they do not prove trusted origin or efficacy.
  • Keep raw evidence separate from sanitized reports; never persist credentials, real secrets, or unnecessary personal data.
  • Preserve blocked and not-tested outcomes; do not downgrade them into clean results.
  • Promotion requires independent review; an operator cannot approve their own result.
  • Link [authorization](../cyber-authorization/SKILL.md) and [mapping](../cyber-framework-mapping/SKILL.md) only when evidenced.

Anti-Patterns

  • **No scanner-only finding**: Corroborate tool output or mark suspected.
  • **No invented certainty**: Use limitations and status fields.
  • **No unredacted persistence**: Minimize and sanitize before storage.
  • **No rewritten history**: Append corrections with provenance.

References

  • [Evidence record fields](references/evidence-record.md)
  • [Canonical security review conventions](../../common/common-security-audit/references/trust-review-policy.md)
Read more
Ships withagent-skills-standard

The portable SDLC standards layer for AI coding agents. Sync once, then work in your own runtime.

Get the whole plugin

Other skills on agent-skills-standard.