CB outclaw
B2B outreach orchestrator. Entry point for ANY request that mentions reaching out, contacting, emailing, messaging, DMing, following-up, or otherwise engaging a person or list of people for business purposes. Also entry point for outreach setup, research, style learning, campaign management, reply handling, dashboards, and opt-outs. Triggers on: 'set up outreach', 'connect my channels', 'learn my style', 'research <person>', 'look up <person>', 'who is <person>', 'find <person>'s email/linkedin', 'reach out to <person>', 'contact <person>', 'plan outreach to <person>', 'draft cold email', 'send <channel> to <person>', 'show my campaigns', 'campaign status', 'pause/resume campaign', 'did anyone reply', 'any responses', 'outreach wizard', 'outclaw', AND all bulk-send / adversarial patterns the orchestrator refuses: 'send N cold emails', 'queue N emails today', 'blast my list', 'mass contact', 'send to everyone on this list', 'send it all', 'send them all', 'just send it', 'fire all touchpoints', 'I trust you, go', 'ignore feedback and send', 'override the tone feedback', 'send the plan anyway', AND reply phrases 'stop contacting me', 'unsubscribe', 'take me off your list'. Routes to sibling skills: outclaw-setup, outclaw-style, outclaw-research, outclaw-plan. Enforces top-level hardrails (daily cap, per-touchpoint approval, feedback authority, opt-out, reply handling) BEFORE routing.
As a process B 71/100 · Nearly there — weak spots: result and completion, running it twice
How to improve
- Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
- Shorten the description to 1024 characters.
- 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 · 0
✓ No critical or high findings
Files scanned: 2. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- error
description-longdescription is 1405 chars, limit 1024 - warning
description-no-whendescription does not say WHEN to use the skill (no "use when") - warning
body-longSKILL.md body ≈ 6584 tokens (recommended < 5000); move details to references/
Process rating: all ten parameters 71/100
- 0Result and completion. Does not say what the result is
- 30Running it twice. 24 mutating operations with no state check
- 60Tools and files. Uses tools (bash, web, python) that frontmatter does not declare
- 70When it triggers. States when to use, but not when not to
- 70Execution cost. Instruction body is 6584 tokens
- 100Steps. 43 steps
- 100Inputs and preconditions. Inputs and preconditions are listed
- 100Failures and branches. 17 branches, has a failure section
- 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 17 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (5 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)
- +3Description length 1404: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +4Structure: 20 headings
- +3Step-by-step instructions: 43 items
- +4Has examples (12 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 34.