CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

research-deep

Read research outline, launch independent agent for each item for deep research. Disable task output.

Quality

68%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./skills/research-codex-zh/research-deep/SKILL.md

The canonical home for this skill is research-deep in Weizhena/Deep-Research-skills

SKILL.md
Quality
Evals
Security

Quality

Content

67%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 skill body is a well-sequenced, actionable batch-research workflow with concrete prompt and validation artifacts. Its main weakness is a redundant one-shot example that inflates length and an implicit error-recovery loop.

Suggestions

Replace the full One-shot Example with a brief 2-3 line variable substitution note, since it duplicates the prompt template.

Make the validate->fix->retry loop explicit ('If validation fails, review errors, fix the JSON, re-run validation; only then is the item complete').

Specify how to launch the web-search-agent (concrete command or invocation) and state defaults for batch_size and items_per_agent.

DimensionReasoningScore

Conciseness

The body is mostly efficient with no concept over-explanation, but the full 'One-shot Example' largely duplicates the prompt template with concrete values, adding padding that could be trimmed.

3 / 5

Actionability

It provides an executable, copy-paste prompt template, an exact validation CLI command, and a slugify rule; minor gaps remain in the agent-launch mechanism (no concrete command) and the outline.yaml schema is referenced but not defined.

4 / 5

Workflow Clarity

A clear 5-step sequence with explicit validation ('validate_json.py', 'Task is complete only after validation passes') and resume/approval checkpoints is present, but the validate->fix->retry recovery loop on validation failure is not explicitly laid out, keeping it below a 5.

4 / 5

Progressive Disclosure

The single file is well-organized into Trigger, Workflow, and Agent Config sections with no nested references; since there are no bundle files to split, structure is appropriate, though the inline template+example detail is borderline dense for a 5.

4 / 5

Total

15

/

20

Passed

Description

53%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 states concrete actions and a recognizable niche but omits any explicit usage trigger, which caps its completeness and limits trigger-term coverage. It is moderately specific and distinctive but not self-activating.

Suggestions

Add a 'Use when...' clause naming natural trigger phrases (e.g., 'Use when running deep research from an outline.yaml, gathering sources for each item, or doing parallel multi-topic research').

Broaden trigger terms with synonyms users actually say (investigate, literature/source review, parallel research) rather than relying only on 'research'.

Drop or contextualize 'Disable task output' in the description, as it is an implementation detail that does not aid triggering or distinctiveness.

DimensionReasoningScore

Specificity

It names the domain ('deep research') and two concrete actions ('Read research outline', 'launch independent agent for each item'), matching the 'domain + 1-2 concrete actions but not comprehensive' anchor; it does not enumerate research outputs or downstream steps.

3 / 5

Completeness

It has a clear 'what' (read outline, launch per-item agents) but no 'Use when...' clause or equivalent trigger guidance, so per the boundary guidance completeness is capped at 3.

3 / 5

Trigger Term Quality

It surfaces 'research' and 'deep research' as natural terms, but misses common synonyms and variations a user might say (investigate, literature review, gather sources), fitting the 'some relevant keywords but missing variations' anchor.

3 / 5

Distinctiveness Conflict Risk

The parallel-agent-per-outline-item niche is fairly distinctive and unlikely to fire for unrelated skills, with only minor overlap risk against other research skills; it is above 'somewhat specific' but lacks the comprehensive trigger phrasing of a 5.

4 / 5

Total

13

/

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

Weizhena/Deep-Research-skills

Path

skills/research-codex-zh/research-deep/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.