BC deep-research
Universal deep research with 13-agent pipeline on Hermes Agent. 7 modes: full research, quick brief, paper review, lit-review, fact-check, Socratic guided research dialogue, systematic review with meta-analysis. Uses delegate_task for each agent. Triggers on: research, deep research, literature review, systematic review, meta-analysis, PRISMA, evidence synthesis, fact-check, guide my research, help me think through, 研究, 深度研究, 文獻回顧, 系統性回顧, 後設分析, 事實查核, 引導我的研究, 幫我釐清.
Universal deep research with 13-agent pipeline on Hermes Agent.
As a process C 59/100 · Has gaps — weak spots: inputs and preconditions, consistency, progress reporting
How to improve
- Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 3
✓ No critical or high findings
Medium and low: 3
-
low Risky intent
intent-offensive-securityreferences/interdisciplinary_bridges.md:164Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)| Computer Science | Achieving fairness and explainability at the technical level | Fairness metrics, XAI, alignment | Algorithm design, benchmarking | Bias detection tools, explainable models, red te
detector -
low Exfiltration
exfil-secret-in-urlshared/cross_model_verification.md:191Credential passed in a URL query string (normal for some APIs — verify the host is the intended service) (placeholder value)curl -s "https://generativelanguage.googleapis.com/v1beta/models/${ARS_CROSS_MODEL}:generateContent?key=…" \placeholder -
low Exfiltration
net-credential-useshared/cross_model_verification.md:191Credential used in a network call (verify the destination is the intended service) (destination is a well-known publishing service)curl -s "https://generativelanguage.googleapis.com/v1beta/models/${ARS_CROSS_MODEL}:generateContent?key=…" \known service
Files scanned: 58. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
description-no-whendescription does not say WHEN to use the skill (no "use when")
Process rating: all ten parameters 59/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Progress reporting. Says nothing while it works
- 40Consistency. Frontmatter name (deep-research) differs from the folder (ars-deep-research)
- 50Failures and branches. 0 branches, has a failure section
- 60Tools and files. Uses tools (bash) that frontmatter does not declare
- 60Result and completion. Output format stated, no completion criterion
- 70When it triggers. States when to use, but not when not to
- 85Steps. 35 steps, 1 vague phrases
- 100Execution cost. Instruction body is 3514 tokens
- 100Running it twice. No mutating operations
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 19 top-level sections: this looks like several domains in one skill
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)
- -212 emoji in the instructions: noise for the model
- +2Single-language instructions
- +3Description length 468: enough signal without eating the budget
- +4Structure: 28 headings
- +3Step-by-step instructions: 35 items
- +3Output format is stated explicitly
- +4Has examples (13 code blocks)
- +4Reference files are cited in the instructions (12 of 20)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 78.