BC skill-hub-query
Query, install, update, and edit AI agent skills on any compatible Skill Hub (self-hosted, or any Hub implementing the documented API contract). Dual-channel: with a token it uses the authenticated API (full features including private skills); without a token it falls back to the unauthenticated channel (public skills only). Covers: list newly published skills, search by keyword / author / time / source, inspect version history, install or upgrade a specific version, and edit a skill's card metadata (display name, summary, tags, visibility, applicable position, etc.) via a safety-first GET -> diff -> backup -> PUT -> dual-channel verify -> auto-rollback flow. Trigger phrases include "what's new on the hub", "search for X skill", "install X", "update Y skill", "edit hub card info", "show skill version history", "skill-hub-query".
Query, install, update, and edit AI agent skills on any compatible Skill Hub (self-hosted, or any Hub implementing the documented API contract).
As a process C 57/100 · Has gaps — weak spots: result and completion, when it triggers
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.
How to improve
- 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 · 5
✓ No critical or high findings
Medium and low: 5
-
medium Exfiltration
net-credential-usescripts/_lib.sh:558Credential used in a network call (verify the destination is the intended service)curl_opts+=(-H "${hdr}: ${scheme}${token}") -
medium Exfiltration
net-credential-usescripts/_lib.sh:659Credential used in a network call (verify the destination is the intended service)curl_opts+=(-H "${hdr}: ${scheme}${token}") -
medium Dangerous commands
cmd-shell-rcSKILL.md:158Writes to a shell startup fileecho 'export SKILL_HUB_URL="https://hub.your-company.com"' >> ~/.bashrc
-
medium Dangerous commands
cmd-shell-rcSKILL.md:159Writes to a shell startup fileecho 'export SKILL_HUB_TOKEN="<your-token>"' >> ~/.bashrc
-
medium Dangerous commands
cmd-shell-rcSKILL.md:192Writes to a shell startup fileecho 'export SKILL_HUB_PROVIDER=skillhub_cn' >> ~/.bashrc
Files scanned: 14. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
body-longSKILL.md body ≈ 6118 tokens (recommended < 5000); move details to references/
Process rating: all ten parameters 57/100
- 0Result and completion. Does not say what the result is
- 20When it triggers. No condition that starts the skill
- 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
- 70Execution cost. Instruction body is 6118 tokens
- 85Steps. 29 steps, 1 vague phrases
- 100Consistency. Name and required fields are in place
- 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 14 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (16 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
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 840: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +5Description quotes 7 example trigger phrases
- +4Structure: 32 headings
- +3Step-by-step instructions: 29 items
- +4Has examples (13 code blocks)
- +4Reference files are cited in the instructions (1 of 1)
- +3All 8 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 83.