CtrlK
BlogDocsLog inGet started
Tessl Logo

scienceworld-target-identifier

Analyzes room observations to identify objects matching a given target description (e.g., 'living thing'). Triggered after exploring a room when the agent needs to locate a specific type of item. Processes the observation list, filters objects based on target criteria, and returns candidate objects for further action.

59

Quality

67%

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

Fix and improve this skill with Tessl

tessl review fix ./experiments/src/skills/scienceworld/scienceworld-target-identifier/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

50%

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

The skill body is well-structured with a clear workflow and a useful worked example, but its actionability and progressive disclosure suffer from referencing a bundled classification script that does not exist and an empty action_patterns.md reference file.

Suggestions

Either add the referenced 'bundled classification script' to the scripts/ bundle or replace Step 2's reliance on it with concrete, self-contained filtering logic the agent can execute directly.

Replace vague 'bundled resources'/'reference taxonomy' mentions with explicit markdown links (e.g., See [taxonomy.md](references/taxonomy.md)) so navigation is clearly signaled.

Either populate references/action_patterns.md with its promised content or remove it from the bundle to avoid an empty stub.

DimensionReasoningScore

Conciseness

The body is well-organized and mostly efficient without explaining concepts Claude already knows, but the Key Considerations section repeats information from the workflow and references a non-existent 'bundled classification script', so it could be tightened rather than being fully lean.

2 / 3

Actionability

The Example Application gives concrete, copy-paste-ready commands (look around, focus on turtle egg, pick up turtle egg, teleport to bathroom), but the core Step 2 instruction to 'Use the bundled classification script' points to a script that does not exist in the bundle, leaving a key step non-executable.

2 / 3

Workflow Clarity

The four-step sequence (Parse, Filter, Prioritize, Generate Action Plan) plus Error Handling recovery branches is clear, but validation checkpoints are only implicit (no explicit 'confirm match before acting' gate) and the missing classification script introduces ambiguity.

2 / 3

Progressive Disclosure

Bundle references are one level deep (references/taxonomy.md is real and substantial), but the body never links them with markdown navigation, references a 'bundled classification script' with no corresponding file, and references/action_patterns.md is an empty stub, so structure and signaling are incomplete.

2 / 3

Total

8

/

12

Passed

Description

85%

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 specific, complete, and clearly scoped to a distinct niche with an explicit trigger clause. Its main weakness is trigger-term quality, which relies on domain jargon and omits common user variations.

Suggestions

Add more natural-language trigger variations users might actually say (e.g., 'find a living thing', 'locate an object', 'what can I pick up here') alongside the technical phrasing.

Consider including the core ScienceWorld action that triggers the skill (e.g., 'after running look around') more prominently as a natural keyword.

DimensionReasoningScore

Specificity

Names multiple concrete actions — 'Analyzes room observations to identify objects', 'Processes the observation list, filters objects based on target criteria', 'returns candidate objects' — matching the anchor that lists several specific concrete actions rather than a single vague action.

3 / 3

Completeness

Clearly answers both what (analyzes/processes/filters/returns candidates) and when ('Triggered after exploring a room when the agent needs to locate a specific type of item') with an explicit trigger clause equivalent to a 'Use when' clause, so it is not capped at 2.

3 / 3

Trigger Term Quality

Contains relevant trigger terms ('Triggered after exploring a room', 'locate a specific type of item', e.g. 'living thing') but leans on environment jargon ('room observations', 'observation list') and lacks common user-spoken variations, so it matches 'some relevant keywords but missing common variations' rather than full coverage.

2 / 3

Distinctiveness Conflict Risk

Occupies a clear niche — ScienceWorld target-object identification triggered after look around — with distinct, environment-specific triggers unlikely to conflict with other skills.

3 / 3

Total

11

/

12

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.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
zjunlp/SkillNet
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.