From tsuga
One-shot procedure for turning a raw incident dump (Slack threads, incident reports, GitHub PRs, Tsuga CLI output) into a populated `skills/incident-history/references/incidents/` archive with one folder per incident, each containing a SUMMARY.md with a validated `## Diagnostic path` section. Trigger this skill when bootstrapping `incident-history` from scratch for a new deployment, refreshing an existing archive with new incidents, or reformatting an incident tracker's export into the shape your investigation runtime expects. Inputs: `inputs/incidents/<INC-id>/` directories holding raw material per incident. Outputs: `skills/incident-history/references/incidents/<INC-id>/SUMMARY.md` + `_inventory.csv`.
How this skill is triggered — by the user, by Claude, or both
Slash command
/tsuga:build-incident-historyThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Procedure for bootstrapping the `incident-history` skill from raw incident material.
Procedure for bootstrapping the incident-history skill from raw incident material.
skills/incident-history/references/incidents/ populated with one folder per incident:
skills/incident-history/references/incidents/
├── _inventory.csv ← index: incident-id, title, declared_at, last_iso, service, team, severity
├── INC-0001/
│ ├── SUMMARY.md ← the canonical ~300-line dossier
│ └── metadata.json ← {incident_id, declared_at, last_iso, title, severity}
├── INC-0002/
│ └── …
└── …
SUMMARY.md is the load-bearing artifact. Everything downstream (the $incident-history skill's analogue-search behavior, your investigation runtime's retrieval, the snapshot-filter in any entrypoint.sh) reads from it. metadata.json exists so a SNAPSHOT_AT filter can drop future incidents without parsing prose.
## Diagnostic path commands have gone stale after a CLI change.references/INPUT_LAYOUT.md — what raw material you need, where to put it, what each source contributes.references/PROCEDURE.md — phase-by-phase workflow. Follow sequentially.references/SUMMARY_TEMPLATE.md — the canonical SUMMARY.md shape with exemplar section content.references/LESSONS.md — the gotchas that will bite you if you skip them.references/SUBAGENT_PROMPT.md — exact prompt template for the per-incident fan-out.references/VERIFICATION.md — the acceptance gates. Every incident folder must pass before shipping.INC-id, one raw-input directory, one output path.## Diagnostic path section is what downstream agents actually read for analogue search. Every command in it must parse and execute against a real tsuga CLI. No MCP-tool pseudo-syntax. No rtk prefix.metadata.json with at minimum declared_at and last_iso (ISO 8601) is required for the snapshot-filter. Incidents missing it get silently dropped by entrypoint.sh.tsuga command in their Diagnostic path sections. If any command fails to parse, fix the template and re-run the subagent batch. Do not hand-patch individual files.Provides UI/UX resources: 50+ styles, color palettes, font pairings, guidelines, charts for web/mobile across React, Next.js, Vue, Svelte, Tailwind, React Native, Flutter. Aids planning, building, reviewing interfaces.
Fetches up-to-date documentation from Context7 for libraries and frameworks like React, Next.js, Prisma. Use for setup questions, API references, and code examples.
npx claudepluginhub tsuga-dev/agent-plugins --plugin tsuga