BB review-responder
Use this skill when an operator is actively running the Google Business Profile review-response workflow for one of their configured client accounts. Specific triggers: 'check for new reviews,' 'run the review check for [client],' 'new review came in for [client],' 'draft a reply to the [reviewer] review,' 'approve the draft for [reviewer/client],' 'post the reply for [review id],' 'show pending review approvals,' 'show me the pending reviews,' or 'apply the [medical/legal/restaurant/retail/general] industry profile to this draft.' Do NOT trigger on: general questions about how to handle reviews, casual mentions of Google reviews, marketing strategy chat, requests to write a review (versus reply to one), or any workflow where no configured client exists. Covers: scheduled checks against Google Business Profile API for configured clients, star-rating-matched draft replies with industry-aware drafting constraints (medical/HIPAA-aware, legal, restaurant, retail), and an approval gate across Telegram, email, webhook, or in-chat channels. Drafts are NEVER auto-posted; operator approval is required before any reply is published. See Privacy and Data Handling for credential and posting scope.
As a process B 68/100 · Nearly there — weak spots: inputs and preconditions, consistency
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.
- 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 · 2
✓ No critical or high findings
Medium and low: 2
-
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
-
low Exfiltration
exfil-send-secrets-to-urlSKILL.md:286Instruction to send secrets/history to an external endpoint (negated — the text forbids it)- Per-client OAuth credentials (`oauth_client_id`, `oauth_client_secret`, `refresh_token`) live in JSON files under `clients_dir`. These are the operator's credentials for the operator's own clients.
negated
Files scanned: 10. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- error
description-longdescription is 1204 chars, limit 1024
Process rating: all ten parameters 68/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 40Consistency. Frontmatter name (review-responder) differs from the folder (google-review-responder)
- 60Tools and files. Uses tools (web, python) 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
- 100Steps. 73 steps
- 100Failures and branches. 3 branches, has a failure section
- 100Execution cost. Instruction body is 3539 tokens
- 100Running it twice. Mutating operations check current state
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 12 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)
- +3Description length 1204: 120–800 characters recommended
- +1No license
- +2Single-language instructions
- +4Structure: 29 headings
- +3Step-by-step instructions: 73 items
- +3Output format is stated explicitly
- +4Has examples (7 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 59.