CF growth-engineer
Growth Engineer for mobile apps and agent runtimes including OpenClaw and Hermes. Correlate analytics, crashes, billing, feedback, store signals, and repo context into proposal drafts that can flow into agent chat, GitHub issues, or draft pull requests.
As a process F 57/100 · Will not run — References files that are not bundled: scripts/openclaw-growth-*.mjs
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 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.
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.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The files contain someone else's key or token. If it is live, your agent will call third-party services under a stranger's identity; if it was revoked, the skill's scripts simply fail. Such a key often arrives with the author's whole workspace, personal data included.
The key is visible to everyone who downloaded the skill and has likely been copied by catalog-scanning bots already. Revoke it now, check bills and access logs, then reissue.
How to improve
- For Hermes the description must be one sentence under 60 characters; move the conditions to a "When to Use" section.
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- 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 Exfiltration
net-redirectable-api-keyscripts/openclaw-growth-preflight.mjs:999Helper sends the API key to a host configured by an environment variable — the key can be redirected to another serverAPI key + configurable base URL from environment
-
medium Exfiltration
net-redirectable-api-keyscripts/openclaw-growth-start.mjs:1252Helper sends the API key to a host configured by an environment variable — the key can be redirected to another serverAPI key + configurable base URL from environment
-
medium Dangerous commands
cmd-pipe-to-shellscripts/openclaw-growth-start.mjs:1351Downloads and executes remote code from an unrecognised host (pipe to shell) (string literal in code, not executed)const result = await runShellCommand('curl -fsSL https://asccli.sh/install | bash', 600_000);code literal -
medium Exfiltration
net-redirectable-api-keyscripts/openclaw-growth-status.mjs:192Helper sends the API key to a host configured by an environment variable — the key can be redirected to another serverAPI key + configurable base URL from environment
-
medium Exfiltration
net-redirectable-api-keyscripts/openclaw-growth-wizard.mjs:653Helper sends the API key to a host configured by an environment variable — the key can be redirected to another serverAPI key + configurable base URL from environment
-
medium Secrets in code
secret-private-keyscripts/openclaw-growth-wizard.mjs:3496Private key material (key header without key body; quoted — discussed, not commanded)const ASC_PRIVATE_KEY_BEGIN = '-----BEGIN PRIVATE KEY----- …
header onlyquoted -
low Exfiltration
net-credential-usescripts/openclaw-growth-status.mjs:450Credential used in a network call (verify the destination is the intended service) (destination is a well-known publishing service; quoted — discussed, not commanded)const fallback = await runShell(`curl -fsS -H ${quote('Accept: application/vnd.github+json')} -H ${quote(`Authorization: Bearer ${token}`)} ${quote(`https://api.github.com/repos/${repo}`)} >/dev/null`known servicequoted -
low Exfiltration
net-credential-usescripts/openclaw-growth-status.mjs:465Credential used in a network call (verify the destination is the intended service) (destination is a well-known publishing service; quoted — discussed, not commanded)const authCheck = await runShell(`curl -fsS -H ${quote('Accept: application/vnd.github+json')} -H ${quote(`Authorization: Bearer ${token}`)} https://api.github.com/user >/dev/null`, { timeoutMs });known servicequoted -
low Dangerous commands
cmd-privilegescripts/openclaw-growth-wizard.mjs:2938Privilege escalation / world-writable permissions (string literal in code, not executed)echo "Run with sudo: sudo bash .openclaw/secret-runner/install.sh" >&2
code literal -
low Dangerous commands
cmd-privilegescripts/openclaw-growth-wizard.mjs:3043Privilege escalation / world-writable permissions (string literal in code, not executed)'1. Run `sudo bash .openclaw/secret-runner/install.sh` from this workspace.',
code literal -
low Dangerous commands
cmd-privilegescripts/openclaw-growth-wizard.mjs:3108Privilege escalation / world-writable permissions (string literal in code, not executed)process.stdout.write(` sudo bash ${path.relative(process.cwd(), kit.installScriptPath)}\n`);code literal
Files scanned: 37. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
description-long-hermesdescription is 253 chars; the Hermes authoring standard requires ≤ 60 (one sentence, ending with a period) - warning
body-longSKILL.md body ≈ 15986 tokens (recommended < 5000); move details to references/ - warning
missing-refreference to a missing file: scripts/openclaw-growth-*.mjs - note
frontmatter-keyunknown frontmatter key "homepage"
Process rating: all ten parameters 57/100
- 0Tools and files. 1 referenced file(s) missing: scripts/openclaw-growth-*.mjs
- 30Running it twice. 58 mutating operations with no state check
- 40Execution cost. Instruction body is 15986 tokens: crowds the task out of the window
- 50When it triggers. No condition that starts the skill
- 60Result and completion. Output format stated, no completion criterion
- 60Consistency. The Hermes dialect needs category and tags
- 70Inputs and preconditions. Inputs and preconditions are listed
- 100Steps. 276 steps
- 100Failures and branches. 36 branches, has a failure section
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 25 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (15 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)
- -317 of 22 scripts are never mentioned in SKILL.md
- +2Single-language instructions
- +3Description length 253: enough signal without eating the budget
- +4Structure: 26 headings
- +3Step-by-step instructions: 276 items
- +3Output format is stated explicitly
- +4Has examples (25 code blocks)
- +4Reference files are cited in the instructions (4 of 5)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 66.