BB deep-research
Deep research skill — broad parallel web searches, multi-source validation, confidence tracking, cited Markdown report. Supports 11 research types: market (TAM/SAM, segments, pricing, trends), domain (industry structure, ecosystem, regulatory landscape), technical (architecture, tools, benchmarks), competitive (competitor teardown, positioning, win/loss), product (feature analysis, reviews, roadmap signals), academic (literature survey, citation networks, key authors), person/org (due diligence on a company or public figure), financial (funding rounds, valuation multiples, revenue signals), legal (IP, patents, litigation, compliance), trend (emerging signals, foresight, scenario mapping), community (ecosystem health, key voices, governance, fragmentation). Use when asked to: 'research <topic>', 'deep dive on X', 'analyze the landscape', 'competitive analysis', 'compare these options', 'who are the players in Z', 'literature review', 'background on Y', 'what papers exist on X', 'product teardown', 'technology evaluation', 'regulatory overview', 'funding landscape', 'what trends are emerging in X', 'patent landscape', 'community health', or any request requiring scanning many sources and producing a cited written analysis. Apply whenever the deliverable is a thorough, sourced report rather than a quick answer. Trigger even when phrased casually: 'look into X', 'what's the deal with Y', 'dig into Z', 'I need to understand the space', 'catch me up on X'.
As a process B 68/100 · Nearly there — weak spots: result and completion, inputs and preconditions
What is at stake
Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The skill asks for more than the task needs: broad tool access, credential environment variables, binaries. Every extra permission widens the damage from a mistake or a compromise.
Narrow allowed-tools and the variable list to the minimum; replace binaries with readable sources or scripts.
How to improve
- Shorten the description to 1024 characters.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 1
✓ No critical or high findings
Medium and low: 1
-
medium Broad scope
meta-broad-allowed-toolsSKILL.md:1Broad tool permissions pre-approved: Bash(curl:*)allowed-tools: Read Edit Write Glob Grep Agent WebFetch WebSearch AskUserQuestion Bash(curl:*) Bash(pandoc:*) Bash(md-to-pdf:*)
Files scanned: 16. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- error
description-longdescription is 1474 chars, limit 1024
Process rating: all ten parameters 68/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 70When it triggers. States when to use, but not when not to
- 70Failures and branches. 11 branches
- 100Tools and files. Tools declared in frontmatter
- 100Steps. 39 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 2738 tokens
- 100Running it twice. Mutating operations check current state
- 100Progress reporting. Reports progress
- low 11 top-level sections: this looks like several domains in one skill
- medium 6 test cases, all positive: not one "should refuse" or "should ask first"
- low No test case covers injection arriving through data
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 1474: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +2Single-language instructions
- +4Structure: 11 headings
- +3Step-by-step instructions: 39 items
- +4Has examples (3 code blocks)
- +4Reference files are cited in the instructions (13 of 13)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 61.