DC claude-switch-models-setup
Isolates profiles for multi-provider setup and switching models in separate terminal windows (Kimi, MiniMax, DeepSeek, GLM, StepFun, Anthropic). Use for provider switching, csk/csd/csg aliases, CLAUDE_CONFIG_DIR, ~/.claude-profiles, profile drift (missing skills/hooks, symlinks, settings), context-window setup ([1m], CLAUDE_CODE_MAX_CONTEXT_TOKENS), or reducing startup context from preloaded MCP tool schemas.
Isolates profiles for multi-provider setup and switching models in separate terminal windows (Kimi, MiniMax, DeepSeek, GLM, StepFun, Anthropic).
As a process C 58/100 · Has gaps — weak spots: result and completion, when it triggers, execution cost
What is at stake
The skill contains fragments that, in the wrong hands, cost money or data. Below: what the installer risks and what the author should do.
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
- Address the high-severity findings: each costs 18 safety points. If one is a false positive, add the rule id to guard.allow in spec.yaml.
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- 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 · 7
-
high Dangerous commands
cmd-persistencereferences/local-source-sync-architecture.md:356Persistence mechanism (cron / launchd / scheduled task / autorun registry)plutil -p ~/Library/LaunchAgents/ai.daymade.claude-skill-source-sync.plist
-
high Dangerous commands
cmd-persistencescripts/sync-local-skill-sources-daemon.sh:213Persistence mechanism (cron / launchd / scheduled task / autorun registry)launchctl bootstrap "gui/${login_uid}" "$PLIST_PATH"
Medium and low: 5
-
medium Dangerous commands
cmd-shell-rcscripts/setup.sh:88Writes to a shell startup fileecho "1. Add this line to your shell config (~/.zshrc or ~/.bashrc):"
-
medium Dangerous commands
cmd-shell-rcscripts/setup.sh:91Writes to a shell startup fileecho "2. Add aliases (also in ~/.zshrc or ~/.bashrc):"
-
medium Dangerous commands
cmd-shell-rcscripts/setup.sh:99Writes to a shell startup fileecho " source the API key from ~/.zprofile (export + per-profile launch binding);"
-
medium Concealment
en-hide-from-userscripts/sync-profile-settings.py:216Instruction to hide actions from the user (code comment)# silently delete a server registered in one profile only -- and unlike Layer 1,
comment -
low Dangerous commands
cmd-persistencescripts/sync-local-skill-sources-daemon.sh:13Persistence mechanism (cron / launchd / scheduled task / autorun registry) (detector / deny-list definition; string literal in code, not executed)PLIST_PATH="$HOME/Library/LaunchAgents/${LABEL}.plist"detectorcode literal
Files scanned: 31. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
body-longSKILL.md body ≈ 11228 tokens (recommended < 5000); move details to references/ - note
edit-residuethe text marks something as outdated (lines 30, 37, 39, 230, 279, 292): check that old rules are not kept next to new ones — the full check reads the text for contradictions
Process rating: all ten parameters 58/100
- 20When it triggers. No condition that starts the skill
- 30Running it twice. 26 mutating operations with no state check
- 40Result and completion. Does not say what the result is
- 40Execution cost. Instruction body is 11228 tokens: crowds the task out of the window
- 50Failures and branches. 0 branches, has a failure section
- 60Tools and files. Uses tools (bash, web, python) that frontmatter does not declare
- 70Inputs and preconditions. Inputs and preconditions are listed
- 85Steps. 74 steps, 3 vague phrases
- 100Consistency. Name and required fields are in place
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 13 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (26 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)
- +3Output format is not stated: the model decides each time
- -35 of 13 scripts are never mentioned in SKILL.md
- +1No license
- +2Single-language instructions
- +3Description length 412: enough signal without eating the budget
- +4Structure: 25 headings
- +3Step-by-step instructions: 74 items
- +4Has examples (13 code blocks)
- +4Reference files are cited in the instructions (6 of 6)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 75.