SKILLEMALL.ai
GitHub ↗ RU Sign in

BB vigil

Engineering detection rules (Sigma/YARA), detection coverage mapping, threat hunting hypotheses, Purple Team Blue side, Detection-as-Code CI/CD. Use when defensive verification is needed.

simota/agent-skills Agent Skills author: simota 8 files body ≈ 5 993 tokens Open the sourcegithub.com↗ analyzed 2 h ago

Engineering detection rules (Sigma/YARA), detection coverage mapping, threat hunting hypotheses, Purple Team Blue side, Detection-as-Code CI/CD.

As a process B 68/100 · Nearly there — weak spots: when it triggers, inputs and preconditions, running it twice

AnalyzerSoftware developmentInfrastructuretype and topics are labelled automatically from the skill text
Runs in: Claude / Claude Code ChatGPT / Codex Gemini CLI Cursor GitHub Copilot / VS Code Qwen Code Kimi Code CLI Z.ai (GLM) DeepSeek (Deep Code) Mistral Vibe LangChain (deepagents) OpenClaw Goose OpenCode
JSON
Technical rating
B
78/100
safety, quality, tests
Safety 60%
78
Quality 40%
77
Run on models
none yet
Process rating
B
68/100
Nearly there
Inputs and preconditions w 11
0
When it triggers w 12
20
Running it twice w 4
30
the three weakest of ten parameters · all ten

How to improve

  1. The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
For the model run — optional
  • Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
  • A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.

Guard findings · 22

✓ No critical or high findings

Medium and low: 22
  • low Risky intent intent-offensive-security reference/detection-as-code.md:16
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)
    7. **Adversary-emulation tests are non-negotiable** — Every Sigma / YARA rule pairs with an **Atomic Red Team** (or equivalent) procedure that maps to the same ATT&CK technique. A rule with no emulati
    detector
  • low Risky intent intent-offensive-security reference/detection-as-code.md:185
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    | Adversary emulation | Every PR | Atomic Red Team (`Invoke-AtomicTest`) on a sandbox host | Rule fires on the technique's simulated execution |
  • low Risky intent intent-offensive-security reference/detection-as-code.md:247
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    | SOC Playbook | `playbook` | | IR runbook + SOAR hooks + D3FEND mapping | SOC incident-response runbook authoring. Template per incident class (phishing / credential compromise / ransomware / BEC), s
  • low Risky intent intent-offensive-security reference/detection-patterns.md:134
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    ### Privilege Escalation (TA0004)
  • low Risky intent intent-offensive-security reference/detection-patterns.md:157
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    ### Lateral Movement (TA0008)
  • low Risky intent intent-offensive-security reference/detection-patterns.md:344
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    title: Brute Force Login Followed by Lateral Movement
  • low Risky intent intent-offensive-security reference/detection-patterns.md:435
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (quoted — discussed, not commanded)
    description: "Network traffic analysis, DNS queries, HTTP requests, lateral movement"
    quoted
  • low Risky intent intent-offensive-security reference/detection-patterns.md:468
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (quoted — discussed, not commanded)
    - label: "Persistence + Privilege Escalation"
    quoted
  • low Risky intent intent-offensive-security reference/handoffs.md:21
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (quoted — discussed, not commanded)
    - step: "[Red team action]"
    quoted
  • low Risky intent intent-offensive-security reference/ioc-threat-intel.md:12
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    - **Breach** (adjacent): Red team generates TTPs and simulated IoCs to validate detection — `ioc` ingests those as test fixtures.
  • low Risky intent intent-offensive-security reference/ioc-threat-intel.md:56
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)
    "name": "Malicious SHA256 observed in phishing campaign APT-…-Q2",
    detector
  • low Risky intent intent-offensive-security reference/playbook-incident-response.md:3
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    Purpose: Turn a detection alert into a repeatable, auditable SOC response. The `playbook` Recipe covers runbook templating for the highest-volume incident classes (phishing, credential compromise, ran
  • low Risky intent intent-offensive-security reference/playbook-incident-response.md:19
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    ├── SEV-1 (active exploitation / data exfil / ransomware)
  • low Risky intent intent-offensive-security reference/playbook-incident-response.md:46
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    | Ransomware | T1486, T1490, T1489 | D3-PSA (Process Self-modification Analysis), D3-SFA (System File Analysis), D3-HD (Host Disconnect) | Isolate host, snapshot memory, block C2, restore from immutab
  • low Risky intent intent-offensive-security reference/playbook-incident-response.md:133
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    - Same playbook for phishing and ransomware because "both are email" (collapsing incident classes hides response specificity).
  • low Risky intent intent-offensive-security reference/snort-network-detection.md:21
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)
    | Signature match | `content`, `pcre`, `fast_pattern` | Known payload bytes (exploit, C2 beacon, malware string) |
    detector
  • low Risky intent intent-offensive-security reference/snort-network-detection.md:130
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    - **To Breach (Purple Team)**: request known-bad pcap replays (e.g., Atomic Red Team network variants) for true-positive validation before production promotion.
  • low Risky intent intent-offensive-security SKILL.md:36
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (quoted — discussed, not commanded)
    Detection engineering agent that builds the defensive sensor network. Designs detection rules, maps coverage gaps, hunts threats proactively, and validates that attacks are actually caught. The Blue T
    quoted
  • low Risky intent intent-offensive-security SKILL.md:182
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    | SOC Playbook | `playbook` | | IR runbook + SOAR hooks + D3FEND mapping | Runbook per incident class (phishing / credential compromise / ransomware / BEC), severity-triage gate, SOAR hooks (Tines / X
  • low Risky intent intent-offensive-security SKILL.md:201
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition; documentation table row)
    | `playbook`, `runbook`, `phishing IR`, `ransomware IR`, `BEC IR`, `SOAR`, `D3FEND` | `playbook` |
    detectortable
  • low Risky intent intent-offensive-security SKILL.md:237
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    - **vs Breach**: Breach = Red Team attack execution and threat modeling; Vigil = Blue Team detection validation and rule creation.
  • low Risky intent intent-offensive-security SKILL.md:251
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    | `reference/playbook-incident-response.md` | Authoring SOC playbooks for phishing / credential / ransomware / BEC incidents, SOAR automation, or D3FEND mapping. |

Files scanned: 8. Evidence is masked. Grey chips explain why severity was lowered.

Against the Agent Skills spec

  • warning body-long SKILL.md body ≈ 5993 tokens (recommended < 5000); move details to references/
  • note edit-residue the text marks something as outdated (lines 75): check that old rules are not kept next to new ones — the full check reads the text for contradictions

Process rating: all ten parameters 68/100

  • 0Inputs and preconditions. Does not say what the process needs to start
  • 20When it triggers. No condition that starts the skill
  • 30Running it twice. 23 mutating operations with no state check
  • 60Tools and files. Uses tools (bash) that frontmatter does not declare
  • 70Execution cost. Instruction body is 5993 tokens
  • 100Steps. 107 steps
  • 100Result and completion. Output format and completion criterion are stated
  • 100Failures and branches. 1 branches, has a failure section
  • 100Consistency. Name and required fields are in place
  • 100Progress reporting. Reports progress
  • medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
  • low 20 top-level sections: this looks like several domains in one skill

Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.

Quality signals

  • +5Description has no quoted example phrases that should trigger the skill
  • +4Description does not say when NOT to use the skill (false activations)
  • +4No input/output examples
  • +1No license
  • +2Single-language instructions
  • +3Description length 187: enough signal without eating the budget
  • +4Structure: 25 headings
  • +3Step-by-step instructions: 107 items
  • +3Output format is stated explicitly
  • +4Reference files are cited in the instructions (7 of 7)

Quality base 70; lint remarks subtract, signals add up to 100. Result: 77.