BF agency-agents
Activate specialized AI agent personas for targeted expertise. The Agency provides pre-built personas across 18+ domains including engineering, marketing, product, sales, design, and specialized roles. Use when you need to adopt a focused professional identity with proven workflows, communication style, and domain expertise. Common triggers: "Activate Frontend Developer mode", "Switch to Product Manager", "Use Data Analyst persona", or when working on domain-specific tasks that benefit from specialized perspective and proven best practices.
As a process F 31/100 · Will not run — References files that are not bundled: references/agents/<domain>/, references/agents/product/, references/agents/sales/
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 contains commands that delete files, rewrite disks or execute code fetched from the network. The agent may run them without asking if it believes the instructions require it.
Replace destructive commands with safe equivalents that ask for confirmation, scope them to one folder, and stop piping curl into a shell: pin a version and a checksum.
How to improve
- The text references files that are not there: add them or drop the references.
- 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 · 11
✓ No critical or high findings
Medium and low: 11
-
medium Dangerous commands
cmd-autorun-instructionreferences/agents/game-development/unreal-engine/unreal-systems-engineer.md:58Instructs the agent to auto-run a script on every session- Always run `GenerateProjectFiles.bat` after modifying `.Build.cs` or `.uproject` files
-
low Risky intent
intent-offensive-securityreferences/agents/engineering/engineering-security-engineer.md:28Offensive-security / dual-use content (legitimate for authorised testing; review intended use)### Vulnerability Assessment & Penetration Testing
-
low Risky intent
intent-offensive-securityreferences/agents/engineering/engineering-solidity-smart-contract-engineer.md:485Offensive-security / dual-use content (legitimate for authorised testing; review intended use)- When access control gaps allow privilege escalation through role chaining
-
low Risky intent
intent-offensive-securityreferences/agents/engineering/engineering-threat-detection-engineer.md:32Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)- Validate that detections actually fire by running atomic red team tests or purple team exercises
detector -
low Risky intent
intent-offensive-securityreferences/agents/engineering/engineering-threat-detection-engineer.md:58Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)- Cover the full kill chain — detecting only initial access means you miss lateral movement, persistence, and exfiltration
detector -
low Risky intent
intent-offensive-securityreferences/agents/engineering/engineering-threat-detection-engineer.md:184Offensive-security / dual-use content (legitimate for authorised testing; review intended use)| Privilege Escalation| 13 | 5 | 8 | 38% |
-
low Risky intent
intent-offensive-securityreferences/agents/engineering/engineering-threat-detection-engineer.md:188Offensive-security / dual-use content (legitimate for authorised testing; review intended use)| Lateral Movement | 9 | 4 | 5 | 44% |
-
low Risky intent
intent-offensive-securityreferences/agents/engineering/engineering-threat-detection-engineer.md:191Offensive-security / dual-use content (legitimate for authorised testing; review intended use)| Command and Control | 16 | 5 | 11 | 31% |
-
low Secrets in code
secret-high-entropy-tokenreferences/agents/engineering/engineering-threat-detection-engineer.md:234High-entropy token-like string (may be an id, hash or a credential)run: pip install sigma-cli pySigma-backend-splunk pySi…der
-
low Secrets in code
secret-high-entropy-tokenreferences/agents/engineering/engineering-threat-detection-engineer.md:273High-entropy token-like string (may be an id, hash or a credential)pySi…der \
-
low Secrets in code
secret-high-entropy-tokenreferences/agents/game-development/unreal-engine/unreal-multiplayer-architect.md:299High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)- Implement `URep…n2D` for open-world games: only replicate actors within spatial cells to nearby clients
quoted
Files scanned: 80. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
frontmatter-yamlSKILL.md: the frontmatter is not valid YAML (YAML parse error: Nested mappings are not allowed in compact mappings at line 2, column 14: description: Activate specialized AI agent personas for targeted expertise. The… ^ ); fields were read line by line. The usual cause is a colon inside an unquoted value - warning
missing-refreference to a missing file: references/agents/<domain>/ - warning
missing-refreference to a missing file: references/agents/product/ - warning
missing-refreference to a missing file: references/agents/sales/
Process rating: all ten parameters 31/100
- 0Tools and files. 3 referenced file(s) missing: references/agents/<domain>/, references/agents/product/, references/agents/sales/
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Failures and branches. Linear process with no failure handling
- 0Progress reporting. Says nothing while it works
- 20When it triggers. No condition that starts the skill
- 40Consistency. Frontmatter name (agency-agents) differs from the folder (agency-agents-wrapper)
- 100Steps. 45 steps
- 100Execution cost. Instruction body is 1297 tokens
- 100Running it twice. No mutating operations
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
- +4Description does not say when NOT to use the skill (false activations)
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +5Description quotes 3 example trigger phrases
- +3Description length 546: enough signal without eating the budget
- +4Structure: 10 headings
- +3Step-by-step instructions: 45 items
- +4Has examples (2 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 67.