Skip to content

reply-team/jason-ai / skills/business/sdr-operations

sdr-operations

Verified

FileKindSize
SKILL.mdskill md31.7 KB
lexicon/neutral.yamlconfig20.8 KB
operations/01-discovery.yamlconfig6.5 KB
operations/02-contacts.yamlconfig9.9 KB
operations/03-audiences.yamlconfig6.9 KB
operations/04-import.yamlconfig2.2 KB
operations/05-campaigns.yamlconfig15.5 KB
operations/06-steps.yamlconfig10.2 KB
operations/07-messages.yamlconfig3.9 KB
operations/08-enrollment.yamlconfig5.3 KB
operations/09-conversations.yamlconfig4.5 KB
operations/10-tasks.yamlconfig5.0 KB
operations/11-meetings.yamlconfig9.6 KB
operations/12-signals.yamlconfig4.8 KB
operations/13-inbound.yamlconfig5.7 KB
operations/14-privacy.yamlconfig12.0 KB
operations/15-sending.yamlconfig18.0 KB
operations/16-social.yamlconfig6.6 KB
operations/17-oversight.yamlconfig16.1 KB
operations/18-team.yamlconfig6.7 KB
operations/19-accounts.yamlconfig6.2 KB
operations/20-measurement.yamlconfig5.1 KB
operations/21-introspection.yamlconfig8.1 KB
operations/families.yamlconfig15.5 KB
references/catalog-01-discovery.mdmarkdown5.6 KB
references/catalog-02-contacts.mdmarkdown7.1 KB
references/catalog-03-audiences.mdmarkdown5.2 KB
references/catalog-04-import.mdmarkdown2.5 KB
references/catalog-05-campaigns.mdmarkdown11.4 KB
references/catalog-06-steps.mdmarkdown7.6 KB
references/catalog-07-messages.mdmarkdown3.8 KB
references/catalog-08-enrollment.mdmarkdown5.0 KB
references/catalog-09-conversations.mdmarkdown4.1 KB
references/catalog-10-tasks.mdmarkdown4.2 KB
references/catalog-11-meetings.mdmarkdown7.5 KB
references/catalog-12-signals.mdmarkdown4.2 KB
references/catalog-13-inbound.mdmarkdown4.9 KB
references/catalog-14-privacy.mdmarkdown9.2 KB
references/catalog-15-sending.mdmarkdown13.0 KB
references/catalog-16-social.mdmarkdown5.3 KB
references/catalog-17-oversight.mdmarkdown11.2 KB
references/catalog-18-team.mdmarkdown5.4 KB
references/catalog-19-accounts.mdmarkdown5.2 KB
references/catalog-20-measurement.mdmarkdown4.7 KB
references/catalog-21-introspection.mdmarkdown6.6 KB
references/chains.mdmarkdown63.3 KB
references/invariants.mdmarkdown39.7 KB
references/non-operations.mdmarkdown14.2 KB
references/open-questions.mdmarkdown40.9 KB
references/operation-index.mdmarkdown41.4 KB

references/catalog-05-campaigns.md

View on GitHub
<!-- Generated. Do not edit by hand. -->

# Family 5 — Campaign, schedule, windows and limits

Generated from the contract fragments in `operations/`. Edit a fragment, not this file.

*The programme* · 26 operations · contract 2.0.0

**Reading the table.** † marks a core operation. The five properties are *reach*
(`read` changes nothing · `control` changes state we own · `act` reaches the outside world),
*reversibility* (`reversible` · `compensatable` · `irreversible`), *approval*
(`auto` · `confirm_once` · `confirm_each`, derived from reversibility × reach), *before repeating*
(the observable state to read before running it again) and *cost* (`none` · `metered`, with its
basis and the meter it consumes). *Key* and *per-item* are the two standing obligations: an
idempotency key on every act, every metered call, every write accepting a collection and every
write creating a durable object. A write that takes no key says why in *exempt* — `absolute_valued`
(the same value written again), `terminal_state` (the same state arrived at again), or `unstated`,
which is the contract admitting it has no reason and not a claim that one exists. And one outcome
per item whenever a collection is carried. Where a property is conditional the cell holds the
dangerous reading, with the condition beneath it.

| Operation | Intent | Reach | Reversibility | Approval | Before repeating | Cost | Key | Per-item | Invariants |
|---|---|---|---|---|---|---|---|---|---|
| `campaign.list` † | "What is running?" — every campaign with its state and health: fine, not progressing, degraded, blocked. | `read` | `reversible` | `auto` | nothing at stake | `none` | `none` | `not_applicable` | — |
| `campaign.get` † | "Show me this campaign" — settings, senders, schedule and timezone policy, pacing, reply policy, declared successor, and the count of people in each state. | `read` | `reversible` | `auto` | nothing at stake | `none` | `none` | `not_applicable` | — |
| `campaign.create` | "Start a new campaign" — always a draft; creating never sends and never arms anything. | `control` | `reversible` | `auto` | list campaigns for one already created under this key | `none` | `required` | `not_applicable` | `E1` |
| `campaign.clone` | "Build this one from the one that worked" — structure, steps, variants, schedule, pacing and reply policy; never enrolments, history or results. | `control` | `reversible` | `auto` | a campaign already created under this key | `none` | `required` | `not_applicable` | `E1` |
| `campaign.update` | "Change the campaign's own settings" — name, owner, description, personalisation tier, declared successor, and the exit rules that are not reply events. | `control` | `reversible` | `auto` | read the campaign; a setter is last-write-wins | `none` | `none` | `not_applicable` | — |
| `reply_policy.set` | "What happens when they answer?" — one decision per event class, plus the window inside which a reply is still attributed to this campaign. | `control` | `reversible` | `confirm_once`<br>*departs:* raised above the derived auto: turning "stop on reply" off is how a campaign keeps writing to somebody who already answered, which is simultaneously the rudest and the most reputationally damaging thing an outbound programme can do. The operation is safe to repeat; the confirmation is about the content of the decision, not the mechanics of the call. | the policy in force, from `campaign.get` | `none` | `none` | `not_applicable` | `A14`, `G6` |
| `schedule.set` | "When may this campaign send?" — windows per day, timezone policy with its ordered fallback, delay basis, out-of-window behaviour, attached blackout calendars. | `control` | `reversible` | `confirm_once`<br>*departs:* raised above the derived auto, because a schedule change silently rewrites every future date in the campaign. | the schedule in force, and re-resolve the calendar — the same authored delay is a different date under a different basis. | `none` | `none` | `not_applicable` | `E7`, `E8` |
| `pacing.set` | "Slow this campaign without slowing everything else on the same mailbox" — people admitted per day and messages sent per day, plus minimum gap, jitter and cold-start ramp. | `control` | `reversible` | `auto` | the pacing in force, from `campaign.get` | `none` | `none` | `not_applicable` | `F1` |
| `sender_binding.set` | "Send this from these mailboxes, numbers or social accounts" — for a campaign or for named participations. | `act`<br>*conditional:* `control` for new entrants only; `act` where it rebinds live participations, because the prospect sees a new From address mid-thread; resolved by the preview. | `compensatable` | `confirm_once`<br>*artefact:* preview | the current bindings and the per-enrolment rebind ledger | `none` | `required` | `required` | `F9`, `H5`, `H13` |
| `campaign.validate`<br>*questions:* 12 | "Is this safe to turn on?" — missing or unpublished content, steps no sender can serve, a schedule that never opens, no terminal condition, the count already enrolled, projected first-24-hour volume, projected metered cost, whether the test send passed, and any successor campaign. | `read` | `reversible` | `auto` | nothing at stake — validation is a report, and the gate is a human reading it. | `none` | `none` | `not_applicable` | `E2`, `E4`, `E5` |
| `campaign.test` | "Let me receive step 1 myself, and click the unsubscribe link, before two hundred strangers do" — only to addresses the workspace can demonstrate it owns. | `act` | `irreversible` | `confirm_each` | the test ledger for this campaign, step and address; a timed-out test may already have arrived. | `metered`<br>*basis:* per message<br>*meter:* `message` | `required` | `required` | `E5` |
| `campaign.activate` | "Turn it on" | `act`<br>*conditional:* `act`, with magnitude equal to the enrolled population; `control` only where the campaign holds no enrolments, and even then it arms the campaign so the next enrolment sends immediately. | `irreversible` | `confirm_once`<br>*artefact:* preview | the campaign's live state and its first-touch ledger — an activation whose result was lost may already be sending. | `metered`<br>*basis:* per message released<br>*meter:* `message` | `required` | `required`<br>*conditional:* when a collection is given. | `E1`, `E2`, `E3` |
| `campaign.pause` †<br>*questions:* 9 | "Stop this campaign" — incident response: never requires a lookup first and never waits for permission. | `control` | `reversible` | `auto` | the campaign's state; pausing a paused campaign is a no-op, not an error. | `none` | `required`<br>*conditional:* when a collection is given. | `required` | `E6`, `G4`, `L6` |
| `campaign.resume` | "Start it sending again" — refuses unless it is actually paused. | `act` | `irreversible` | `confirm_once`<br>*artefact:* preview | the campaign's state and its first-touch ledger since the attempt. | `metered`<br>*basis:* per message released<br>*meter:* `message` | `required` | `required`<br>*conditional:* when a collection is given. | `E3`, `E6`, `G5`, `L6` |
| `campaign.freeze`<br>*questions:* 9, 25 | "Suspend it over the holidays, the conference, the incident" — carries a mandatory expiry and resumes itself. | `control` | `reversible` | `auto` | the campaign's state and the freeze's expiry — freezing twice with different expiries is a change, not a repeat. | `none` | `none`<br>*exempt:* `unstated` | `required`<br>*conditional:* when a collection is given. | `E6`, `G4` |
| `campaign.unfreeze` | "Let it go early" | `act` | `irreversible` | `confirm_once`<br>*artefact:* preview | the campaign's state and the first-touch ledger since the attempt. | `metered`<br>*basis:* per message released<br>*meter:* `message` | `required` | `required`<br>*conditional:* when a collection is given. | `E3`, `E6` |
| `campaign.archive` | "Retire it" — refused while active enrolments remain unless they are first stopped with a reason. | `control` | `compensatable` | `confirm_once` | the campaign's archived state | `none` | `required`<br>*conditional:* when a collection is given. | `required` | — |
| `campaign.unarchive` | "Bring it back into the working set" — never resumes sending. | `control` | `reversible` | `auto` | the campaign's archived state | `none` | `required`<br>*conditional:* when a collection is given. | `required` | — |
| `schedule.resolve` | "Show me the actual calendar" — day by day, for a campaign or one person's participation, including what a weekend, a holiday, a closed window or a deferral moved. | `read` | `reversible` | `auto` | nothing at stake | `none` | `none` | `required`<br>*conditional:* when a collection of enrolments is given. | `E7` |
| `send_window.resolve` | "When may this person be contacted on this channel?" — legal quiet hours in their local time, operator blackouts, working days, the campaign's windows, and the basis used for each. | `read` | `reversible` | `auto` | nothing at stake | `none` | `none` | `required`<br>*conditional:* when a collection of people is given. | `A20`, `E8` |
| `blackout.define` | "Nothing goes out that week" — named date ranges and region-coded holiday calendars, each with defer-or-skip and whose region applies. | `control` | `reversible` | `confirm_once`<br>*departs:* raised above the derived auto: a blackout suppresses sending across every campaign its region and date range cover, and defer-or-skip decides whether the touches it stops come back at all. Lifting it again is itself a release of suppressed sending, so the range is agreed once, when it is drawn. | list blackouts for one already created under this key | `none` | `required` | `required`<br>*conditional:* when a collection of ranges is given. | `A20`, `E8` |
| `blackout.list` | "What is currently suppressing sending, and until when?" | `read` | `reversible` | `auto` | nothing at stake | `none` | `none` | `not_applicable` | — |
| `blackout.remove` | "Lift the suppression" | `act`<br>*conditional:* `act` where the removal exposes a window open now on a live campaign; `control` otherwise — resolved by the preview. | `irreversible`<br>*conditional:* irreversible in the act case, reversible otherwise. | `confirm_once`<br>*artefact:* preview | the blackout list and the first-touch ledger for the affected campaigns. | `metered`<br>*basis:* per message released<br>*meter:* `message` | `required` | `required` | `A20`, `E3` |
| `frequency_policy.define`<br>*questions:* 11 | "Nobody hears from us more than this often" — max touches per person, account or domain per window type; which message classes count and which are exempt; and whether a breach defers with a date or skips. | `control` | `reversible` | `confirm_once`<br>*departs:* raised above the derived auto: the cap declared here is evaluated for every campaign, every channel and every sender at once, and which message classes count has no default — a policy written loosely either silences the answers people asked for or lets one person hear from us three times in a day. | the policy in force; a second define under the same key must not create a second policy. | `none` | `required` | `not_applicable` | `F2`, `F10` |
| `frequency_policy.get` | "What cap is in force here, and where did it come from?" | `read` | `reversible` | `auto` | nothing at stake | `none` | `none` | `not_applicable` | — |
| `capacity.estimate` | "Can this plan physically run?" — sending capacity, windows, blackouts, pacing and existing load over a stated horizon, against the volume the plan implies. | `read` | `reversible` | `auto` | nothing at stake | `none` | `none` | `not_applicable` | `A6`, `F1`, `F3` |
slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."Stars: 198kUpdated Claude CodeCodexVerifiedInstall
ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt.Stars: 194kUpdated Claude CodeCodexVerifiedInstall