BC space-duck-kimi-relay
Optional Lane A / BYOB add-on for Space Duck — runs a local RFC 8628 device-code "Sign in with Kimi" flow (no password; browser-approved) so a self-hosted duck can use the owner's flat-rate Kimi membership for inference. Credentials (access + rotating refresh token) are stored locally at ~/.kimi-code/credentials/kimi.json (0600) and are NEVER sent to Spaceduckling. Contacts only auth.kimi.com and api.kimi.com; inference is processed by Moonshot AI in China (no Western data residency). Optional pay-per-token fallback to openrouter.ai when OPENROUTER_API_KEY is set (daily-capped). Runs a localhost-only proxy (127.0.0.1, default 8471) protected by an auto-generated 0600 bearer secret. Hosted (Lane B) ducks use the Mission Control card instead. Triggers on "sign in with kimi", "kimi membership login", "clawhub space-duck kimi", "kimi relay login".
Optional Lane A / BYOB add-on for Space Duck — runs a local RFC 8628 device-code "Sign in with Kimi" flow (no password; browser-approved) so a self-hosted…
As a process C 61/100 · Has gaps — weak spots: result and completion, inputs and preconditions, running it twice
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.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The skill tells the agent to hide things from you: not to show errors, not to mention actions, to report differently from what was done. You lose the ability to see what the agent really did.
Transparency beats a smooth answer. If the goal is to hide technical noise, ask the agent to "summarise briefly", not to "not mention".
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 · 4
✓ No critical or high findings
Medium and low: 4
-
medium Dangerous commands
cmd-persistencescripts/kimi_login.py:448Persistence mechanism (cron / launchd / scheduled task / autorun registry) (string literal in code, not executed)print("launchctl load failed: " + (r.stderr or r.stdout), file=sys.stderr)code literal -
medium Dangerous commands
cmd-persistencescripts/kimi_login.py:496Persistence mechanism (cron / launchd / scheduled task / autorun registry) (string literal in code, not executed)plist = os.path.expanduser("~/Library/LaunchAgents/com.spaceduckling.kimi-relay.plist")code literal -
medium Dangerous commands
cmd-persistencescripts/kimi_login.py:546Persistence mechanism (cron / launchd / scheduled task / autorun registry) (string literal in code, not executed)plist = os.path.expanduser("~/Library/LaunchAgents/com.spaceduckling.kimi-relay.plist")code literal -
medium Concealment
en-hide-from-userSKILL.md:91Instruction to hide actions from the user (quoted — discussed, not commanded)membership can't silently run up a pay-per-token bill; past the cap the
quoted
Files scanned: 5. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- note
frontmatter-keyunknown frontmatter key "disclosures"
Process rating: all ten parameters 61/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 30Running it twice. 1 mutating operations with no state check
- 60Tools and files. Uses tools (bash, web) that frontmatter does not declare
- 70When it triggers. States when to use, but not when not to
- 100Steps. 12 steps
- 100Failures and branches. 1 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 1473 tokens
- 100Progress reporting. Reports progress
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)
- +3Description length 855: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- -2localhost URLs: will not work for another user
- +1No license
- +2Single-language instructions
- +5Description quotes 5 example trigger phrases
- +4Structure: 8 headings
- +3Step-by-step instructions: 12 items
- +4Has examples (2 code blocks)
- +3All 1 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 86.