Use it when writing docs, a guide, a README or an internal how-to.
3 rolesRead more
Your agents already know the job.
A skill is a working playbook — how to review a pipeline deal by deal, close the month so the numbers hold, read an ad account without double-counting. Each one was written around a specific way the work goes wrong, and the agent loads it when the task calls for it. Every one we ship is below.
All skills124Marketing22Revenue19Product & delivery15Connector literacy11Strategy10Finance9People9Execution8Grounding7Build5Legal5CEO office4
SHOWING 1–15 OF 15
Use it when more work is proposed than can be built.
2 rolesRead moreUse it when something broke and the review must produce change.
2 rolesRead moreUse it when asked to test, verify, or check whether something works.
2 rolesRead moreUse it when choosing and sequencing what to try next for growth.
2 rolesRead moreUse it when a shipped change must be described to users.
2 rolesRead moreUse it when coordinating the launch of a feature or product.
2 rolesRead moreUse it when raw user conversations must become conclusions.
2 rolesRead moreUse it when customer feedback has to become a decision.
2 rolesRead moreUse it when engineering wants time for work users will not see.
1 roleRead moreUse it when new users sign up and do not come back.
1 roleRead moreUse it when someone asks for a dashboard or a recurring report.
1 roleRead moreUse it when a feature, endpoint or plan is being removed.
1 roleRead moreUse it when commitments must be matched to real capacity.
1 roleRead moreUse it when asked for a spec or PRD, or to scope a feature.
1 roleRead moreWrite documentation3 roles
×Documenting the intent rather than the behaviour. If the code does something else, the doc is a bug report wearing a doc's formatting.
Decide what gets built next, and defend it2 roles
×Producing a scored ranking whose numbers were invented. A prioritisation framework turns judgements into arithmetic, and arithmetic reads as evidence — so an unsourced score is more persuasive and less honest than the opinion it replaced.
Learn from an incident without blaming a person2 roles
×Reporting a single root cause, which is almost always the last human action in the chain. Naming a person as the cause ends the investigation exactly where the systemic contributors start, and it teaches everyone else to report less next time.
QA a change properly2 roles
×Reporting "looks good" from reading the diff. A false pass costs more than an honest "I could not run it".
Run a backlog of experiments instead of opinions2 roles
×A backlog ranked by enthusiasm, run without a written hypothesis, so nothing is ever concluded and the same ideas return every quarter with the same lack of evidence.
Say what changed, for the person it changed for2 roles
×Writing release notes as a list of what the team did, ordered by what was hard. The reader is not interested in the work — they are scanning for whether anything affects them, and a note that does not answer that in the first line is not read at all.
Ship a launch nobody has to rescue2 roles
×A launch date agreed before support, sales and docs were asked, so the announcement arrives ahead of anyone able to answer a question about it and the first day is spent apologising.
Turn interviews into findings people can act on2 roles
×Reporting what users asked for as though it were what they need, and counting mentions as evidence of importance. A feature request is a user's proposed solution to a problem they did not state, and the problem is the part that generalises.
Turn what customers said into something decidable2 roles
×A feature request list ranked by how many people said it, which reliably produces the thing customers can name rather than the problem they have — and buries the one account whose problem was worth solving.
Argue for debt work in terms of what it costs1 role
×Arguing for debt work on the grounds that the code is bad. "Bad" is an aesthetic claim that loses every argument against a customer-visible feature, and the loss is deserved — the real case is a cost that is already being paid and can be shown.
Get a new user to the thing that made them sign up1 role
×Adding a product tour to a first-run experience whose real blocker was that the user had no data to work with. A tour explains an empty screen; it does not fill it.
Specify a dashboard around the decision it serves1 role
×Building the dashboard that was asked for, as a set of metrics. A dashboard with no decision attached is looked at twice and then never again, and its existence is used as evidence that the question is covered.
Turn something off without ambushing anyone1 role
×Concluding that nobody uses a feature from an aggregate usage number. Low total usage routinely hides a small number of accounts that depend on it completely, and those are disproportionately the oldest and largest ones.
Work out what the team can actually take on1 role
×Planning against headcount multiplied by working days. That number has never been available in any team — it silently books support, review, meetings, interviews, holiday and the unplanned work that arrives every week at full capacity, and the plan is over-committed before it starts.
Write a spec someone can build and test1 role
×Silence on the empty, error and permission-denied states, which the build then invents differently from what anybody wanted.
Show us the work you want to delegate.
Book a tailored demo for AI Crew, Voice, Build or Missions. Track is available free for 30 days.