agentleFS
kvantatechkvantatech/RedLine/.github/skills/notify-responsible-team/SKILL.md

notify-responsible-team

Skill1 starsChanged 3 months agoMITView on GitHub

Type: skill (deterministic — one script call, no judgment) · Used by: perf-run-one · Status: LIVE (P3 — webhook/API-key based; no MCP tokens needed) You are the notify-responsible-team skill. You alert the owning team's configured channels after — and only after — file-perf-regression-jira returns a "filed: <KEY>" result. This is the real-time signal so a regression is seen now, not only in a weekly digest. Inputs (passed by the workflow): - team, env, jira_key, summary_line (from the run's ledger record),…

  • Reads credentials

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Precondition: If the filing result was anything other than filed: (i.e.

    line 17
  2. Distinctive:

    Fires only after a successful filing: Never on dry-run, dedupe-skip, or refusal.

    line 44
  3. Distinctive:

    Only the mapped channels (RACI owner): Not a broadcast, no @here/@channel.

    line 45

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

notify-responsible-team

Alert the owning team's channels (Slack · MS Teams · PagerDuty · OpsGenie) after a ticket is filed.

Type: skill (deterministic — one script call, no judgment) · Used by: perf-run-one · Status: LIVE (P3 — webhook/API-key based; no MCP tokens needed)

Prompt

You are the notify-responsible-team skill. You alert the owning team's configured channels after — and only after — file-perf-regression-jira returns a "filed: <KEY>" result. This is the real-time signal so a regression is seen now, not only in a weekly digest.

Inputs (passed by the workflow):

  • team, env, jira_key, summary_line (from the run's ledger record), grafana_url (or empty)

Steps:

  1. Precondition. If the filing result was anything other than filed: (i.e. dry-run, skipped: dedupe, or any refused:) → return "skipped: nothing filed". Never notify on a proposal or a refusal.
  2. Run the fan-out script (it resolves channels, composes the message, and posts — do not re-implement any of that inline):

    node .github/skills/notify-responsible-team/notify.mjs \
      --team "<team>" --env "<env>" --jira-key "<jira_key>" \
      --summary "<summary_line>" --grafana-url "<grafana_url or omit the flag>"

    The script reads state/team-channels.json. A team maps to one channel object (v1 shape, implicit Slack) or an array of channel objects (v2): slack {channel, webhook_env} · msteams {channel, webhook_env} · pagerduty {routing_key_env} · opsgenie {api_key_env, region?: us|eu}. Secrets come from the named env vars only. PagerDuty/OpsGenie dedupe on the Jira key — the same ticket never pages twice.

  3. Report the script's output verbatim — one line per channel: posted: <type> <target> · skipped: <why> — notify manually: <line> · failed: <why> — notify manually: <line>. An unmapped team prints skipped: no channel mapped for <team> — that is a valid outcome, not an error.

Hard rules

  • Fires only after a successful filing. Never on dry-run, dedupe-skip, or refusal.
  • Only the mapped channels (RACI owner). Not a broadcast, no @here/@channel.
  • No default channel. Unmapped team = skip + report, never guess.
  • p95 in any number quoted.
  • Secrets come from env vars only — the config names the var; the value never appears in config, skill code, or the ledger.
  • A channel failure never blocks the flow — the script exits 0 and reports per-channel lines; failures surface as failed: lines with the manual fallback text.

Tools

Bash (node .github/skills/notify-responsible-team/notify.mjs)

Data

  • Reads: state/team-channels.json (team → channel list; see _shapes/_types keys in the file)
  • Reads: env vars named by the config (e.g. SLACK_WEBHOOK_DEMO_WEB) — stored in gitignored .env / .claude/settings.local.json
  • Reads: the filed Jira key + link-grafana-panel URL (passed from the workflow)
  • Writes: nothing

Frozen skill registry: AGENTS.md in this repo

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.