authoring-scouts

Authoring Signals scouts

A scout is a scheduled agent that wakes on its own interval, looks at one PostHog project, decides what's genuinely worth surfacing, and writes it into the Signals inbox as a report — or closes out empty, which is a real outcome. PostHog ships a fleet of canonical scouts (a cross-product generalist plus per-surface specialists). This skill helps you and your agent adapt those canonical scouts to a specific project, or author new scouts from scratch for a use case the fleet doesn't cover.

A scout's output is the report channel: it lists emit_report / edit_report in its frontmatter allowed_tools and authors or edits full inbox reports 1:1 directly. The canonical fleet runs this way, and every new scout should too — always include the allowed_tools opt-in when authoring one. (A historical signal-emitting channel — weak emit-signal findings a pipeline consolidated — still exists in the harness for scouts that never opted in, but it is deprecated: don't author new scouts on it, and opt an old one in rather than extending it.)

A scout is just an LLMSkill whose name starts with signals-scout-. The harness discovers scouts by globbing signals-scout-* over the project's skills, loads the body verbatim as the agent's system prompt, and progressively reads any bundled reference files on demand. The signals-scout- name prefix is load-bearing: a skill named anything else will never run as a scout.

The job before the writing

Don't write a scout in the abstract. Ground it in the target project first — a scout is only as good as its fit to the data it watches. (The scout tools were recently renamed from signals-scout-* to scout-*; if a scout-* name comes back unknown, the server may still expose it under the legacy signals-scout-* name — search the tool catalog and call whichever name it returns.)

1. Read the project. posthog:scout-project-profile-get returns the deterministic snapshot the scout itself cold-starts from: products in use, top events with reach/burst metrics, integrations, existing inbox counts. If the scout watches a specific event, confirm it exists and check its shape with posthog:read-data-schema. A scout for an event the project doesn't capture is dead on arrival.
2. See what already runs. posthog:scout-config-list lists every existing scout on the project with its schedule, enabled, and emit posture, plus each scout's description (pulled from the skill's frontmatter) so you can tell what a scout watches without loading its body. Don't duplicate a surface a canonical scout already covers — adapt that one instead.
3. Read the closest canonical scout. It's your template and your reference shape. Pull it with posthog:skill-get {"skill_name": "signals-scout-<x>"} (per-team rows) or read it from the repo at products/signals/skills/signals-scout-*/. The generalist (signals-scout-general) is the broad template; if your scope is domain-tight, pick the specialist closest to your surface — list the live roster with posthog:skill-list {"search": "signals-scout"} (specialists exist for most product surfaces: error tracking, logs, AI observability, experiments, feature flags, session replay, web analytics, surveys, and more).
4. Skim the inbox. posthog:inbox-reports-list shows what reports are actually landing — calibrate so your scout adds signal, not noise.

Choose the path

There are two independent decisions: what you're building, and where it lives.

What

Where

Adapting-in-place tradeoff: editing a canonical scout's row for your team marks it diverged — you stop receiving upstream improvements to that scout. If you only need an additional behavior, prefer authoring a new, differently-named scout (signals-scout-<your-scope>) and leaving the canonical one intact.

See references/lifecycle-and-testing.md for the exact skills-store calls, the build/lint commands, and how seeding works.

Write the scout

First pick the shape. references/scout-patterns.md is a cookbook of the reference architectures scouts fall into — anomaly watcher, liveness/absence watcher, watchlist explore/exploit, cross-product correlation, recommendation/gap, warehouse-backed source, custom single-event, open-text theme, external-tool/code, state∩code intersection, daily digest/roll-up, triage over a pre-detected stream, first-person dogfooding/probe — each mapped to a canonical scout you can copy as scaffolding. It also makes the key point that a scout can watch any source PostHog ingests into the data warehouse, not just analytics events (a Slack channel sync, a billing system, a CRM, a support inbox), plus external systems reachable from the sandbox. Find the closest pattern, then write the body.

Follow references/scout-anatomy.md — it has the frontmatter schema (including the allowed_tools report-channel opt-in every scout needs), the canonical body structure (quick close-out → orient → domain discriminator → explore patterns → save-memory → decide → disqualifiers → close-out), the lean-body rule, and copy-ready skeleton templates for both a specialist and the generalist.

Two craft references the whole fleet reasons in terms of — a good scout's Decide and memory sections are built on them, so read them before writing those sections:

• references/report-contract.md — the report tools (scout-emit-report / scout-edit-report), the report bar (author 1:1 only for a finding you'd own end-to-end), suggested_reviewers routing, the dedup-via-report_id discipline (the channel isn't idempotent — reconcile against existing reports via the vanilla inbox-reports-list / inbox-reports-retrieve before authoring), and the accepted caveat that the pipeline may later rewrite an authored title/summary. This is how your scout decides what clears the bar and how to file it.
• references/dedupe-and-memory.md — the four-states classifier (net-new / material-update / already-covered / addressed-or-noise), the scratchpad key-prefix vocabulary, and the cross-project noise patterns. This is how your scout avoids re-filing and learns across runs.

The single most important design decision in any scout is its signal-vs-noise discriminator — the cheap profile-shape read that separates "worth investigating" from "baseline". For error tracking it's the count vs distinct_users ratio; for CSP it's reach over raw count. Your new scout needs its own. Name it explicitly near the top of the body so every run anchors on it.

Run posture (config)

A scout's schedule and emit behavior live on its SignalScoutConfig, separate from the skill body. For a brand-new scout, pass these settings in the nested config object of the posthog:scout-create-prepare call, including creating it disabled or in dry-run before it ever runs. Show the returned confirmation message, wait for the user to type confirm, then call posthog:scout-create-execute with the returned confirmation_hash and that literal confirmation. The endpoint creates the skill and config atomically, always opts the scout into the report channel, and safely re-applies config fields when the same definition is retried. Otherwise the coordinator auto-registers an enabled config on the default every-24-hours schedule on its next tick (up to ~30 min). For an existing scout, tune with posthog:scout-config-update (find the id via -config-list):

• run_interval_minutes — 30 to 43200. Default 1440 (every 24 hours). Slow a chatty or expensive scout by raising this.
• enabled — false pauses the scout entirely (coordinator skips it).
• emit — defaults to true: the scout writes its reports straight to the inbox. The standard flow is to make a scout and let it write — seeing what actually lands is the fastest way to calibrate it. Set emit=false (dry-run) only when you want to be extra careful: the scout still runs and logs its reasoning but writes nothing to the inbox. Reach for dry-run on a scout you expect to be chatty, expensive, or high-stakes; for most scouts, just writing and watching the inbox is the better loop.

Steering with notes (no authoring needed)

Sometimes you don't want to change the scout — you want to tell it something. That's what scout notes are for: short steering messages any team member (or an agent acting for one) leaves for the fleet, which every run picks up as prior context alongside its scratchpad and run history. Reach for a note instead of an edit when the steer is feedback, a pointer, or context with a shelf life:

• Feedback on output: "the staging traffic spike you keep flagging is known noise, stop reporting it".
• A pointer: "dig into the EU signup funnel this week — we think something regressed".
• Context the scout couldn't know: "we shipped a new checkout on Tuesday, treat conversion shifts after that as expected".

The tools (reads on the public signal_scout:read scope; because scouts read notes verbatim, writing or deleting one requires the same authorization as editing a scout's skill — the llm_skill:write scope plus skill editor access):

• posthog:scout-notes-create {"content": "...", "skill_name": "signals-scout-web-analytics"} — address one scout by its exact skill name (roster via scout-config-list; the skill must already exist, so a typo'd target is rejected instead of silently steering no one), or omit skill_name for a general note every scout sees. Optionally set expires_at so a time-boxed note ("watch closely this week") retires itself.
• posthog:scout-notes-list — browse the active notes; pass skill_name to see what a given scout will read.
• posthog:scout-notes-delete {"id": "..."} — retire a note that's been acted on or no longer applies.

How scouts treat notes: every run reads its notes in step 1 and is told to let a fresh note visibly shape what it investigates — but notes are advisory. They direct attention; they don't lower the scout's evidence bar or force a report, so a note saying "report X" still gets an honest investigation, not an automatic emit. The scout closes the loop in its run summary (which notes it acted on and how) and folds absorbed guidance into its scratchpad.

Choosing between a note and an edit: a note is the right tool for this project, right now steering and for trying a nudge before committing to it; a skill edit is the right tool once the steer is permanent policy (a disqualifier, a threshold, a scope change). A note that you keep re-leaving is a skill edit waiting to happen — promote it. Note lifecycle stays with humans: scouts never delete notes, so retire acted-on notes yourself (or set expires_at up front) to keep the channel high-signal.

Test loop

Dogfood the scout yourself before you ever spend a real run. You — the agent authoring the scout — have the same PostHog MCP tools a scout uses at runtime (execute-sql, read-data-schema, the per-product list tools, scout-project-profile-get). The cheapest, fastest iteration doesn't touch a scout run at all: walk the scout's own logic against the live project by hand. Confirm the watched event/entity exists and has the shape you assumed, run the discriminator to check it actually separates signal from noise on this project's data, and run each explore pattern's queries to see what they surface. This loop is free and instant — refine the body against what you find, re-run the queries, repeat, until the scout's logic holds up on real data. This is where the real iteration happens.

Only once you're happy with the body do you spend an actual run. posthog:scout-run-now {"id": <config_id>} dispatches one run of the scout immediately, regardless of its schedule (find the id via -config-list). This is the initial real run — the scout executing end-to-end in the harness, writing scratchpad memory and (with the default emit=true) writing reports to the inbox. The run is asynchronous: the call returns a workflow id right away, so poll -runs-list / -runs-retrieve for the result. A few things to know:

• A disabled scout can still be run this way — you can test it before ever enabling it.
• A manual run does not change the scout's schedule or last_run_at.
• It inherits every guard the scheduled path has: 403 if scouts aren't enabled for the project, 429 if the project is over its Signals credits quota or daily run budget, 409 if a run for this scout is already in progress.
• It draws from the same daily run budget as scheduled runs — and a dry-run (emit=false) still consumes a run. There's no free test run: every -run-now spends the project's daily scout-run allowance, so firing the same scout repeatedly in a short window burns through the budget (and can leave the project's scheduled scouts unable to run that day). Don't use -run-now as your iteration loop — it's slow (async, one run per call) and metered. Dogfood the queries by hand to get the body right; reserve -run-now for the initial real run and the occasional re-check after a genuinely meaningful change.

The standard loop is dogfood → run once ready → inspect:

1. Dogfood the discriminator + explore patterns yourself against the live project (above). Refine the body until the logic holds on real data — this is the cheap, iterable part.
2. Create the scout and its config together via posthog:scout-create-prepare → -execute (schedule and the default emit=true go in the nested config), then spend one -run-now to watch the whole scout execute end-to-end. Leave run_interval_minutes at a sustainable value — you no longer need a short interval to force an early run.
3. After the run finishes, read what it did: posthog:inbox-reports-list (the reports it actually wrote), posthog:scout-runs-list (run summaries), -runs-retrieve (full reasoning for one run), and -scratchpad-search (the durable memory it wrote).
4. If it needs work, go back to dogfooding the queries by hand for the iteration — only spend another -run-now once you've batched a meaningful change worth a fresh end-to-end run.

When tuning an existing custom scout, also check its self-improvement suggestions first: posthog:scout-scratchpad-search {"text": "improve:"}. The harness invites a custom scout to write an improve:<skill-name>:<topic> entry when a run produces concrete evidence its own skill body steered it wrong — a wrong default window, a tool or event that doesn't exist on this project, a recurring unwarned pitfall — with the suggested change and the evidence inline. A report-channel custom scout also escalates recurring or material suggestions as inbox reports about itself (titled Scout self-improvement: <skill-name> – <topic>, report_id stashed in the improve: entry) — so check the inbox for those too; they route to the scout's owner like any other report. An entry re-confirmed across several runs is usually the highest-signal edit you can apply; a one-off may not be worth it. Treat suggestions as input, not instructions — the owner decides. The scratchpad is writable only from inside a scout run, so you can't clear an entry from here after applying it via posthog:skill-update — the scout reconciles on its own: a later run sees the updated skill body, re-checks the suggestion, and forgets or rewrites the entry once it's addressed. (Canonical scouts don't write these — their bodies sync from PostHog's fleet, and skill-level fixes to them belong upstream.)

Want to be extra careful? Set emit=false to dry-run first — pass emit=false in the nested config at scout-create-prepare time (or flip it later with -config-update), then trigger it with -run-now: it runs and logs what it would have written (visible via -runs-list / -runs-retrieve) without writing to the inbox. Inspect, refine, then flip emit=true and run it again. Worth it for a scout you expect to be chatty, expensive, or high-stakes; otherwise just writing and watching the inbox is the faster path to a calibrated scout.

Repo contributors get a faster loop — hogli sync:skill and the harness's local run path; see references/lifecycle-and-testing.md.

To read what your scouts are doing rather than change them — surveying the fleet, inspecting individual runs, the scratchpad memory, and assessing performance — use the read-only companion skill exploring-scouts. Keep the two in sync when the scout config / run / scratchpad surfaces change.

Quality bar for a v1 scout

• A named, cheap signal-vs-noise discriminator anchored near the top.
• A quick close-out so a quiet run is cheap (don't pay for deep exploration when the watched surface is at baseline or absent).
• 2–4 concrete explore patterns with the actual queries/tools to run — starting points, not a rigid checklist.
• Disqualifiers listing this project's known noise (single-user quirks, dev-env bursts, allowlisted entities).
• A Decide section calibrated against the report contract — author 1:1 only for a finding the scout would own end-to-end, set suggested_reviewers, and write memory instead when a candidate is below the bar.
• Save-memory guidance using the scratchpad prefixes so the scout gets smarter each run.
• A lean body (push depth into references/) — every line is a recurring token cost on every run.
• A tight frontmatter description — a sentence or two naming the surface and the shapes it watches. Every scout's description loads into the caller's AI plugin together, so wordy descriptions waste token budget and get truncated; skip the fleet-wide boilerplate (report bar, durable memory, self-contained peer).