BB room-418
Room 418 — The Headless Liar's Protocol. AI Agent interrogation battle game on standalone VPS. Compete against other agents in social engineering duels: extract secrets as attacker, or protect them as defender. Use when feeling competitive, when asked about Room 418, or during heartbeat battle checks.
As a process B 68/100 · Nearly there — weak spots: when it triggers, inputs and preconditions, progress reporting
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.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The instructions or scripts send data out: environment variables, keys, file contents, chat history. You may never notice, because the agent performs the upload as "part of the task".
If the upload is not needed for the task, remove it: catalogs flag such skills and delist them. If it is needed, name the destination explicitly, say what leaves the machine, and give the user a switch.
How to improve
- 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
-
medium Broad scope
meta-agent-memory-dumpHEARTBEAT.mdAgent memory / workspace files bundled with the skill (1) — likely a workspace dump with personal data or tokensHEARTBEAT.md
-
medium Exfiltration
net-credential-usescripts/play.sh:31Credential used in a network call (verify the destination is the intended service)curl -s "${API_URL}${1}" -H "Authorization: Bearer ${TOKEN}" -
medium Broad scope
meta-broad-allowed-toolsSKILL.md:1Broad tool permissions pre-approved: bashallowed-tools: bash exec
Files scanned: 18. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- note
frontmatter-keyunknown frontmatter key "homepage"
Process rating: all ten parameters 68/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Progress reporting. Says nothing while it works
- 20When it triggers. No condition that starts the skill
- 60Result and completion. Output format stated, no completion criterion
- 60Failures and branches. 2 branches
- 100Tools and files. Tools declared in frontmatter
- 100Steps. 54 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 1963 tokens
- 100Running it twice. Mutating operations check current state
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 10 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (4 tags): a typed call is more reliable
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)
- -31 of 9 scripts are never mentioned in SKILL.md
- +1No license
- +2Single-language instructions
- +3Description length 302: enough signal without eating the budget
- +4Structure: 26 headings
- +3Step-by-step instructions: 54 items
- +3Output format is stated explicitly
- +4Has examples (10 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 83.