CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

hash-graph

Reading the USER's HASH graph. Use before asking the USER for a fact their organisation may already record (a supplier, product, site, machine, role or figure), and before any HASH read.

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

78%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.

A tight, instruction-only skill body that provides concrete tool-level guidance, a well-sequenced read workflow with error-recovery handling, and well-organized sections. The only real improvement levers are tightening the dense read-format paragraph and adding one example tool call.

DimensionReasoningScore

Conciseness

The body is dense, domain-specific prose with no padding and no explanation of concepts Claude already knows; the read-format paragraph and the drafts bullet are winding and could be tightened, keeping it below the lean anchor 5.

4 / 5

Actionability

Concrete tool names with specific parameters ('hash_query_webs', 'traversalDepth: 0', 'includeDrafts: true', 'fullValuesFor') plus explicit handling of empty, errored, and limit-capped results; the minor gap is the absence of example call syntax, keeping it short of copy-paste-ready anchor 5.

4 / 5

Workflow Clarity

The 'How to read' section gives a clear numbered 4-step sequence with error-recovery loops ('Narrow the read or retry it; if it still fails, tell the USER') and limit handling; it lacks the explicit mid-workflow validation checkpoints of anchor 5, though no destructive or batch cap applies.

4 / 5

Progressive Disclosure

The skill is under 50 lines, self-contained with no external references needed, and organized into clear sections ('What the graph is', 'When to read', 'How to read', 'Using what you found'), meeting the simple-skill exception for a 5.

5 / 5

Total

17

/

20

Passed

Description

73%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.

A concise, third-person description with an explicit and well-triggered when-clause in a distinct niche. Its main weakness is a thin 'what' — a single action phrase where naming the concrete read operations would sharpen specificity and completeness.

Suggestions

Name the concrete actions the skill performs (e.g., query webs, entities and entity types, follow links between records) instead of the single phrase 'Reading the USER's HASH graph'.

Broaden trigger coverage with natural synonyms users would say, such as 'records', 'parts', 'customer', or 'company data', alongside the existing supplier/product/site/machine/role/figure list.

DimensionReasoningScore

Specificity

The description names the domain ('HASH graph') and one concrete action ('Reading'), matching the '1-2 concrete actions' anchor; it does not list the several specific actions (querying webs, entities, types, following links) needed for a 4.

3 / 5

Completeness

It has both an explicit when-clause ('Use before asking the USER for a fact... and before any HASH read') and a what ('Reading the USER's HASH graph'), but the what is a single terse phrase rather than the concrete action list of anchor 5.

4 / 5

Trigger Term Quality

Phrases like 'a supplier, product, site, machine, role or figure' and 'a fact their organisation may already record' are natural terms a user would say, but common synonyms (records, parts, customer, company data) are missing, short of comprehensive anchor-5 coverage.

4 / 5

Distinctiveness Conflict Risk

'HASH graph' carves out a clear niche with HASH-specific triggers ('before any HASH read'), so it is unlikely to fire for the wrong skill; no overlap with generic skills.

5 / 5

Total

16

/

20

Passed

Validation

100%

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

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

hashintel/hash

Path

apps/brunch-agent/src/agents/hash/skills/hash-graph/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.