CtrlK
BlogDocsLog inGet started
Tessl Logo

tech-search

Self-contained deep tech research. WebSearch + WebFetch + Haiku workers. Pipeline: Query > Decompose > Parallel Search (Haiku) > Evaluate > Synthesize > Document. Zero external dependencies. MCPs optional. Salva em docs/research/{YYYY-MM-DD}-{slug}/.

55

Quality

61%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

High

Do not use without reviewing

Fix and improve this skill with Tessl

tessl review fix ./.claude/skills/tech-search/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

63%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 a well-sequenced, highly actionable research pipeline with strong validation and feedback loops. Its weaknesses are repeated prohibition statements that hurt conciseness and a monolithic inline structure with no reference files to aid progressive disclosure.

Suggestions

Consolidate the repeated prohibitions into a single 'constraints' section and reference it elsewhere instead of restating 'NEVER write outside docs/research/' and 'NEVER implement code' four times each.

Extract the full worker prompt template and the output file manifest into reference files (e.g., references/worker-prompt.md, references/output-structure.md) and link to them from SKILL.md.

Add an explicit post-Document verification step (e.g., confirm all five output files exist) to push workflow_clarity toward 5.

DimensionReasoningScore

Conciseness

The config-style YAML is generally tight, but the same prohibitions are repeated across veto_conditions, constraints, security, and scope_boundaries (e.g., 'NEVER write files outside docs/research/' and 'NEVER implement code' each appear multiple times), so it could be trimmed.

3 / 5

Actionability

Provides a copy-paste-ready worker prompt template, explicit Task() dispatch signature, numeric stopping thresholds, and a concrete output file manifest; only minor gaps keep it from being fully turnkey.

4 / 5

Workflow Clarity

Six phases are clearly sequenced with explicit validation checkpoints (HARD/SOFT stop rules, coverage thresholds, failed-worker fallback, Wave 2 loop), satisfying the batch-operation feedback-loop requirement with only minor implicit checks.

4 / 5

Progressive Disclosure

No bundle files exist and the skill is a single ~430-line monolithic YAML block; content that could live in separate references (worker prompt template, output manifest) is inlined, so structure is present but not split for navigation.

3 / 5

Total

14

/

20

Passed

Description

58%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 concretely names the pipeline stages and tools, giving it solid specificity and distinctiveness. Its main weakness is the absence of any 'when to use' guidance, which caps completeness and leaves trigger-term quality mid-range.

Suggestions

Add an explicit 'Use when ...' clause naming natural user triggers (e.g., 'Use when the user asks to research, investigate, or compare technologies').

Soften jargon by including everyday synonyms users actually say alongside 'deep tech research' (e.g., 'investigate', 'compare', 'look into').

Briefly state the output artifact in user terms (e.g., 'Produces a structured research report') to round out the 'what'.

DimensionReasoningScore

Specificity

Names several concrete actions (Decompose, Parallel Search, Evaluate, Synthesize, Document) and specific tools (WebSearch, WebFetch, Haiku workers, MCPs), with only minor gaps in coverage.

4 / 5

Completeness

Provides a clear 'what' (self-contained research pipeline) but lacks any 'Use when...' or equivalent trigger clause, so per the rubric completeness is capped at 3.

3 / 5

Trigger Term Quality

Has relevant keywords like 'deep tech research' but leans on technical jargon ('Haiku workers', 'MCPs') and misses common natural variations a user would say such as 'investigate', 'compare', or 'look into'.

3 / 5

Distinctiveness Conflict Risk

The explicit multi-phase pipeline and 'self-contained deep tech research' niche make it mostly distinct with only minor overlap risk against general research skills.

4 / 5

Total

14

/

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
SynkraAI/aiox-core
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.