CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

bloodhound-query

Analyze authorized AD collection data in BloodHound CE with the active runtime's supported tools and mirrored GUI.

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

92%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is an exemplary lean operational runbook: every token earns its place, the multi-step workflow carries explicit validation and error-reporting checkpoints, and structure is appropriate for a single-file skill. The only weakness is that the analysis-tool calls stop short of executable argument examples and concrete bound values.

Suggestions

Show example argument shapes for the analysis tools, e.g., bloodhound_mcp_find_objects(name=..., type=...) and bloodhound_mcp_shortest_path(start=..., end=...), so the guidance is copy-paste ready.

Give a concrete bound for row limits and path depth (e.g., "limit results to ~100 rows and max_depth 4") instead of the unspecified "Bound returned rows and path depth".

DimensionReasoningScore

Conciseness

Every sentence delivers non-obvious operational content — exact MCP tool names, ingest status expectations, collection-method constraints, and safety boundaries — with no padding and no explanation of concepts Claude already knows. It leans fully on Claude's competence, matching the anchor-5 example's economy.

5 / 5

Actionability

Concrete named calls with parameters are given for the core flow ("bloodhound_mcp_upload_collection(path=...)", collect with "the matching domain, DC, username, and cache", "DCOnly"/"Default"/"Session") and the analysis tools are all named, but argument shapes for find_objects, object_profile, shortest_path, and cypher_query are unspecified and "Bound returned rows and path depth" gives no concrete values — minor gaps placing it at 4 rather than copy-paste-ready 5.

4 / 5

Workflow Clarity

The hosted engagement is a clearly numbered 1-4 sequence with explicit validation checkpoints: verify the upload's "terminal status is Complete" and report partial/failed ingest, run data_quality and domain_overview after upload, check "GUI follow status before claiming that the operator can see" anything, and report missing collection access rather than inferring safety — with a fallback runtime section for error recovery. This matches anchor 5's explicit validation and feedback loops.

5 / 5

Progressive Disclosure

The body is under 50 lines with no need for external references (tool interfaces live in the MCP schemas), and it is organized into two clearly signaled sections (hosted engagement, standalone OSS runtime) plus a boundary paragraph, which satisfies the simple-skill exception for a top score.

5 / 5

Total

19

/

20

Passed

Description

65%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is distinctive and names its domain precisely, but it lists only a single concrete action and omits any explicit "when to use" trigger guidance, leaving it anchored at the midpoint of the scale. Adding trigger phrases and enumerating the concrete analysis actions would lift both completeness and specificity.

Suggestions

Add an explicit trigger clause, e.g., "Use when the user mentions BloodHound, SharpHound collections, AD attack paths, shortest paths, or Cypher queries against domain data."

Enumerate the concrete actions the skill performs (upload collections, check data quality, find objects, trace shortest paths, run bounded Cypher queries) instead of summarizing them as "Analyze ... collection data".

Include common natural synonyms such as "active directory" spelled out and "attack path" so users' phrasing matches the description.

DimensionReasoningScore

Specificity

The description names the domain ("BloodHound CE", "authorized AD collection data") and one concrete action ("Analyze"), but "the active runtime's supported tools and mirrored GUI" is context rather than additional concrete actions, so it falls short of the several-actions anchor at 4.

3 / 5

Completeness

The "what" is clear (analyze authorized AD collection data in BloodHound CE), but there is no "Use when..." clause or equivalent trigger guidance in the description itself, which caps completeness at 3; the when_to_use string lives in separate metadata, not the description.

3 / 5

Trigger Term Quality

It contains the primary natural terms a user would say — "BloodHound CE", "AD", "collection" — giving good keyword coverage, though natural variations like "active directory" spelled out, "attack path", "shortest path", and "cypher" are missing.

4 / 5

Distinctiveness Conflict Risk

"BloodHound CE" and "AD collection data" carve out a clear niche with distinct triggers and minimal risk of firing for unrelated skills.

5 / 5

Total

15

/

20

Passed

Validation

87%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

Total

14

/

16

Passed

Repository

PurpleAILAB/Decepticon

Path

packages/decepticon/decepticon/skills/standard/ad/bloodhound-query/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.