BC project-manager
Manage projects, grants, milestones, and updates on the Karma protocol. Use when user says "create a project", "new project", "add a grant", "record funding", "add milestone", "complete milestone", "post an update", "project progress", "grant update", "update project", "edit project", "edit grant", "complete grant", "add roadmap milestone", "report impact", "endorse project", "add team member", "set up agent", "configure API key", "check payouts", "payout status", "payout history", "total disbursed", "view invoices", "download invoice", or any project management action.
As a process C 52/100 · Has gaps — weak spots: result and completion, when it triggers, inputs and preconditions
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.
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 · 1
✓ No critical or high findings
Medium and low: 1
-
medium Exfiltration
net-credential-useSKILL.md:543Credential used in a network call (verify the destination is the intended service)curl -s "${BASE_URL}/v2/milestone-invoices/download?key=${INVOICE_FILE_KEY}" \
Files scanned: 2. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
body-longSKILL.md body ≈ 5309 tokens (recommended < 5000); move details to references/
Process rating: all ten parameters 52/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 20When it triggers. No condition that starts the skill
- 40Consistency. Frontmatter name (project-manager) differs from the folder (karma-project-manager)
- 60Tools and files. Uses tools (web) that frontmatter does not declare
- 70Execution cost. Instruction body is 5309 tokens
- 100Steps. 16 steps
- 100Failures and branches. 5 branches, has a failure section
- 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
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)
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +5Description quotes 25 example trigger phrases
- +3Description length 576: enough signal without eating the budget
- +4Structure: 41 headings
- +3Step-by-step instructions: 16 items
- +4Has examples (18 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 79.