@sentry/warden 0.26.2 → 0.27.0

package/CHANGELOG.md

@@ -1,4 +1,240 @@
 # Changelog
+## 0.27.0
+
+### New Features ✨
+
+#### Action
+
+- Support org base configs with repo overlays by @dcramer in [#279](https://github.com/getsentry/warden/pull/279)
+- Write structured findings JSON and expose as output by @gricha in [#233](https://github.com/getsentry/warden/pull/233)
+
+#### Cli
+
+- Build generated skills from path targets by @dcramer in [#282](https://github.com/getsentry/warden/pull/282)
+- Add --staged flag and default to HEAD for uncommitted changes by @dcramer in [#205](https://github.com/getsentry/warden/pull/205)
+- Display confidence level in findings output by @dcramer in [#194](https://github.com/getsentry/warden/pull/194)
+- Always-on JSONL logging with repo-local log directory by @dcramer in [#179](https://github.com/getsentry/warden/pull/179)
+- Add --fail-fast flag to stop after first finding by @dcramer in [#174](https://github.com/getsentry/warden/pull/174)
+- Surface failure details for chunk analysis and finding extraction by @dcramer in [#162](https://github.com/getsentry/warden/pull/162)
+- Interactive fix step-through with y/n/a/s/q prompt by @dcramer in [#161](https://github.com/getsentry/warden/pull/161)
+
+#### Docs
+
+- Improved landing visuals by @PickleNik in [#245](https://github.com/getsentry/warden/pull/245)
+- Add LLM content negotiation and llms.txt by @dcramer in [#178](https://github.com/getsentry/warden/pull/178)
+
+#### Runs
+
+- Write chunk-level JSONL logs by @dcramer in [#265](https://github.com/getsentry/warden/pull/265)
+- Stream in-progress runs via 'warden runs follow' by @dcramer in [#261](https://github.com/getsentry/warden/pull/261)
+
+#### Sdk
+
+- Add finding verification pass by @dcramer in [#290](https://github.com/getsentry/warden/pull/290)
+- Store Claude SDK sessions in .warden/sessions/ by @dcramer in [#193](https://github.com/getsentry/warden/pull/193)
+
+#### Skills
+
+- Add built-in code review skill by @dcramer in [#298](https://github.com/getsentry/warden/pull/298)
+- Add built-in security review skill by @dcramer in [#292](https://github.com/getsentry/warden/pull/292)
+- Consolidate warden-sweep into bundled scripts with prescribed reporting by @dcramer in [#201](https://github.com/getsentry/warden/pull/201)
+- Replace notseer with find-warden-bugs skill by @dcramer in [#200](https://github.com/getsentry/warden/pull/200)
+- Add warden-sweep skill and bundled skill installation by @dcramer in [#196](https://github.com/getsentry/warden/pull/196)
+- Add agent discovery alongside skills by @dcramer in [#136](https://github.com/getsentry/warden/pull/136)
+
+#### Sweep
+
+- Add --skill flag to scan.py by @dcramer in [#211](https://github.com/getsentry/warden/pull/211)
+- Add master tracking issue for sweep PRs by @dcramer in [#210](https://github.com/getsentry/warden/pull/210)
+
+#### Telemetry
+
+- Add skill and model attribution to metrics by @gricha in [#231](https://github.com/getsentry/warden/pull/231)
+- Add namespaced fix-eval feedback metrics by @dcramer in [#229](https://github.com/getsentry/warden/pull/229)
+- Add global scope attributes, run metric, and trace ID surfacing by @dcramer in [#166](https://github.com/getsentry/warden/pull/166)
+- Add input/output attributes to gen_ai spans by @dcramer in [#142](https://github.com/getsentry/warden/pull/142)
+- Add per-turn tracing for Claude Code SDK agent calls by @dcramer in [#141](https://github.com/getsentry/warden/pull/141)
+
+#### Other
+
+- (config) Make auxiliary max retries configurable via defaults.auxiliaryMaxRetries by @dcramer in [#175](https://github.com/getsentry/warden/pull/175)
+- (dedup) Add intra-batch finding consolidation by @dcramer in [#135](https://github.com/getsentry/warden/pull/135)
+- (init) Add section-based output and interactive skill prompt by @dcramer in [#182](https://github.com/getsentry/warden/pull/182)
+- (o11y) Capture run failures in JSONL output by @dcramer in [#259](https://github.com/getsentry/warden/pull/259)
+- (output) Default to COMMENT instead of REQUEST_CHANGES on PRs by @dcramer in [#137](https://github.com/getsentry/warden/pull/137)
+- (runner) Global concurrency pool for file analyses by @dcramer in [#158](https://github.com/getsentry/warden/pull/158)
+- (skill-builder) Build generated repo-local skills by @dcramer in [#280](https://github.com/getsentry/warden/pull/280)
+- Add verification field and confidence filtering by @dcramer in [#191](https://github.com/getsentry/warden/pull/191)
+- Pnpm workspace monorepo with docs migration by @dcramer in [#185](https://github.com/getsentry/warden/pull/185)
+- Adopt dotagents for skill management by @dcramer in [#168](https://github.com/getsentry/warden/pull/168)
+- Cross-location finding merge by @dcramer in [#148](https://github.com/getsentry/warden/pull/148)
+- Add env vars to generated workflow and docs by @dcramer in [#138](https://github.com/getsentry/warden/pull/138)
+- Add Sentry observability (errors, tracing, metrics, structured logs) by @dcramer in [#133](https://github.com/getsentry/warden/pull/133)
+- Split failOn into independent requestChanges and failCheck controls by @dcramer in [#129](https://github.com/getsentry/warden/pull/129)
+
+### Bug Fixes 🐛
+
+#### Action
+
+- Use relative findings output path by @dcramer in [#289](https://github.com/getsentry/warden/pull/289)
+- Handle missing warden.toml gracefully by @gricha in [#235](https://github.com/getsentry/warden/pull/235)
+- Thread maxContextFiles config through Action workflows by @dcramer in [#216](https://github.com/getsentry/warden/pull/216)
+
+#### Cli
+
+- Label built-in skill sources by @dcramer in [#294](https://github.com/getsentry/warden/pull/294)
+- Report final per-file finding counts by @dcramer in [#293](https://github.com/getsentry/warden/pull/293)
+- Align local review diff context by @dcramer in [#276](https://github.com/getsentry/warden/pull/276)
+- Replace process.exit(130) with thrown error for graceful cleanup by @dcramer in [#213](https://github.com/getsentry/warden/pull/213)
+- Write --output file on auth error early exit by @dcramer in [#214](https://github.com/getsentry/warden/pull/214)
+- Normalize path separators in synthetic file changes by @dcramer in [#217](https://github.com/getsentry/warden/pull/217)
+- Exclude node_modules from gitignore file discovery by @dcramer in [#220](https://github.com/getsentry/warden/pull/220)
+- Collapse skipped files into a single count on Ctrl+C by @dcramer in [#172](https://github.com/getsentry/warden/pull/172)
+- Render warnings through Ink's Static component by @dcramer in [#170](https://github.com/getsentry/warden/pull/170)
+- Cap 'Other Files' list in prompts for CLI mode by @dcramer in [#171](https://github.com/getsentry/warden/pull/171)
+- Move completed files to Static to prevent flickering by @dcramer in [#149](https://github.com/getsentry/warden/pull/149)
+- Run all skills locally regardless of trigger type by @gricha in [#145](https://github.com/getsentry/warden/pull/145)
+- Resolve remote config for --skill flag by @dcramer in [#130](https://github.com/getsentry/warden/pull/130)
+
+#### Config
+
+- Allow base configs to use built-in skills by @dcramer in [#305](https://github.com/getsentry/warden/pull/305)
+- Warn on duplicate layered skills by @dcramer in [#304](https://github.com/getsentry/warden/pull/304)
+- Add missing fields to TOML writer for config round-trip by @dcramer in [#218](https://github.com/getsentry/warden/pull/218)
+
+#### Output
+
+- Render GitHub footers as muted text by @dcramer in [#306](https://github.com/getsentry/warden/pull/306)
+- List only successfully applied fixes in PR body by @dcramer in [#219](https://github.com/getsentry/warden/pull/219)
+- Use backticks instead of brackets in attribution footer by @dcramer in [#197](https://github.com/getsentry/warden/pull/197)
+
+#### Sdk
+
+- Short-circuit repeated provider failures by @dcramer in [#291](https://github.com/getsentry/warden/pull/291)
+- Strip invalid suggested fixes from reports by @dcramer in [#228](https://github.com/getsentry/warden/pull/228)
+- Prevent Warden sessions from polluting Claude Code history by @dcramer in [#202](https://github.com/getsentry/warden/pull/202)
+- Handle EPIPE errors from Claude subprocess failures by @dcramer in [#198](https://github.com/getsentry/warden/pull/198)
+- Add verification checklist and Glob tool to reduce analysis hallucinations by @dcramer in [#173](https://github.com/getsentry/warden/pull/173)
+- Constrain skill findings to diff hunk line range by @dcramer in [#151](https://github.com/getsentry/warden/pull/151)
+
+#### Sweep
+
+- Add self-validation to fix subagent prompt by @dcramer in [#223](https://github.com/getsentry/warden/pull/223)
+- Branch worktrees from default branch instead of HEAD by @dcramer in [#221](https://github.com/getsentry/warden/pull/221)
+
+#### Telemetry
+
+- Fill finding analytics gaps by @dcramer in [#311](https://github.com/getsentry/warden/pull/311)
+- Align attributes with OTel semantics by @dcramer in [#307](https://github.com/getsentry/warden/pull/307)
+- Improve OTel GenAI semantic convention compliance by @dcramer in [#140](https://github.com/getsentry/warden/pull/140)
+- Fix missing Sentry traces and negative token costs by @dcramer in [#139](https://github.com/getsentry/warden/pull/139)
+
+#### Other
+
+- (ci) Enable auto-generated changelogs for GitHub releases by @gricha in [#234](https://github.com/getsentry/warden/pull/234)
+- (evals) Update default model to claude-sonnet-4-6 by @dcramer in [#163](https://github.com/getsentry/warden/pull/163)
+- (o11y) Add gen_ai.request.model attribute by @obostjancic in [#247](https://github.com/getsentry/warden/pull/247)
+- (prepare) Skip files with zero content automatically by @dcramer in [#226](https://github.com/getsentry/warden/pull/226)
+- (review) Preserve active Warden threads by @dcramer in [#272](https://github.com/getsentry/warden/pull/272)
+- (sentry) Tag action errors with repository by @dcramer in [#303](https://github.com/getsentry/warden/pull/303)
+- (skill-builder) Stabilize generated skill authoring by @dcramer in [#284](https://github.com/getsentry/warden/pull/284)
+- (skills) Support private repos in `warden add --remote` by @dcramer in [#187](https://github.com/getsentry/warden/pull/187)
+- (usage) Normalize inputTokens to include cached and cache-write tokens by @vgrozdanic in [#143](https://github.com/getsentry/warden/pull/143)
+- Preserve interrupted findings and parse run args by @dcramer in [#297](https://github.com/getsentry/warden/pull/297)
+- Prevent command injection in release workflow by @fix-it-felix-sentry in [#277](https://github.com/getsentry/warden/pull/277)
+- Handle multi-line YAML scalar values in SKILL.md frontmatter by @odysseus0 in [#243](https://github.com/getsentry/warden/pull/243)
+- Server error handler calls process.exit() bypassing cleanup by @dcramer in [#206](https://github.com/getsentry/warden/pull/206)
+- Preserve error chains across error wrapping boundaries by @dcramer in [#215](https://github.com/getsentry/warden/pull/215)
+- Respect ignorePaths in all code paths by @dcramer in [#204](https://github.com/getsentry/warden/pull/204)
+- Exclude dev files and secrets from npm package by @dcramer in [#186](https://github.com/getsentry/warden/pull/186)
+- Clean up orphaned Warden comments when no triggers match by @dcramer in [#134](https://github.com/getsentry/warden/pull/134)
+- Show findings in PR review body when inline comments fail by @dcramer in [#128](https://github.com/getsentry/warden/pull/128)
+- Show skipped status for files not processed during Ctrl+C by @dcramer in [#127](https://github.com/getsentry/warden/pull/127)
+
+### Documentation 📚
+
+- (homepage) Redesign to lead with skills by @dcramer in [#238](https://github.com/getsentry/warden/pull/238)
+- Use built-in security review in samples by @dcramer in [#295](https://github.com/getsentry/warden/pull/295)
+- Use scoped Warden package in npx examples by @dcramer in [#273](https://github.com/getsentry/warden/pull/273)
+- Lead homepage with GitHub PR review story by @dcramer in [#240](https://github.com/getsentry/warden/pull/240)
+- Add org-wide GitHub setup guide by @dcramer in [#239](https://github.com/getsentry/warden/pull/239)
+- Replace non-existent skill names with generic placeholders by @dcramer in [#199](https://github.com/getsentry/warden/pull/199)
+- Simplify getting started guide to use native warden add --remote by @dcramer in [#164](https://github.com/getsentry/warden/pull/164)
+- Fix documentation inconsistencies with codebase by @dcramer in [#147](https://github.com/getsentry/warden/pull/147)
+- Update config reference with missing fields by @dcramer in [#132](https://github.com/getsentry/warden/pull/132)
+
+### Internal Changes 🔧
+
+#### Config
+
+- Use native PR review skills by @dcramer in [#299](https://github.com/getsentry/warden/pull/299)
+- Configure Sentry MCP server by @dcramer in [#301](https://github.com/getsentry/warden/pull/301)
+- Exclude evals directory from warden analysis by @dcramer in [#203](https://github.com/getsentry/warden/pull/203)
+
+#### Deps
+
+- Bump dotagents lib by @gricha in [#296](https://github.com/getsentry/warden/pull/296)
+- Bump yaml from 2.8.2 to 2.8.3 by @dependabot in [#254](https://github.com/getsentry/warden/pull/254)
+
+#### Sdk
+
+- Add flat runtime provider contract by @dcramer in [#270](https://github.com/getsentry/warden/pull/270)
+- Extract applyMergeGroups to unify cross-location merge logic by @dcramer in [#153](https://github.com/getsentry/warden/pull/153)
+
+#### Skills
+
+- Adopt @sentry/dotagents-lib for skill resolution primitives by @gricha in [#285](https://github.com/getsentry/warden/pull/285)
+- Restructure distributed Warden skills by @dcramer in [#268](https://github.com/getsentry/warden/pull/268)
+
+#### Other
+
+- (agents) Update configured skills list by @dcramer in [#241](https://github.com/getsentry/warden/pull/241)
+- (cli) Use git ls-files for .gitignore discovery by @dcramer in [#222](https://github.com/getsentry/warden/pull/222)
+- (dotagents) Gitignore auto-generated files instead of tracking them by @gricha in [#246](https://github.com/getsentry/warden/pull/246)
+- (output) Clean up GitHub finding rendering by @dcramer in [#176](https://github.com/getsentry/warden/pull/176)
+- (skill-builder) Use internal authoring provider by @dcramer in [#283](https://github.com/getsentry/warden/pull/283)
+- (sweep) Clean up warden-sweep skill per skill-creator review by @dcramer in [#225](https://github.com/getsentry/warden/pull/225)
+- Add GitHub Actions workflow skill by @dcramer in [#278](https://github.com/getsentry/warden/pull/278)
+- Add .serena to gitignore by @ericapisani in [#180](https://github.com/getsentry/warden/pull/180)
+- Extract shared skill selection and validate setup-app args by @dcramer in [#208](https://github.com/getsentry/warden/pull/208)
+- Combine multiple reduce operations into single pass by @dcramer in [#207](https://github.com/getsentry/warden/pull/207)
+- Simplify repetitive column width calculations in logs by @dcramer in [#209](https://github.com/getsentry/warden/pull/209)
+- Collapse severity to 3 levels, add warden logs command by @dcramer in [#192](https://github.com/getsentry/warden/pull/192)
+- Migrate to dotagents and add skills reference to AGENTS.md by @dcramer in [#195](https://github.com/getsentry/warden/pull/195)
+- Tighten weak mock assertions in test suite by @dcramer in [#160](https://github.com/getsentry/warden/pull/160)
+- Add conductor.json configuration by @dcramer in [#159](https://github.com/getsentry/warden/pull/159)
+- Remove generated artifacts from dist, keep only ncc bundle by @dcramer in [#131](https://github.com/getsentry/warden/pull/131)
+- Replace batch-based concurrency with sliding window pool by @dcramer in [#126](https://github.com/getsentry/warden/pull/126)
+
+### Other
+
+- release: 0.26.2 by @getsentry-bot in [3d492972](https://github.com/getsentry/warden/commit/3d492972d2e6ae5c2666a5d61d1e37a6591892ff)
+- release: 0.26.1 by @getsentry-bot in [23280325](https://github.com/getsentry/warden/commit/23280325da28b365fb53e6a33b19b167998e5660)
+- release: 0.26.0 by @getsentry-bot in [2bb08b9d](https://github.com/getsentry/warden/commit/2bb08b9d5b276abaee31aab0e672d3ac4ecaee8c)
+- release: 0.25.0 by @getsentry-bot in [56de18fe](https://github.com/getsentry/warden/commit/56de18fee0a1bb6a80deae2c0e4688ab05754126)
+- release: 0.24.1 by @getsentry-bot in [2b9e2b0c](https://github.com/getsentry/warden/commit/2b9e2b0cdab12b950fcb4abc79e33d7db363ee49)
+- release: 0.24.0 by @getsentry-bot in [715259ec](https://github.com/getsentry/warden/commit/715259ecacc9e1e5d78b6b62623960d86e934162)
+- release: 0.23.0 by @getsentry-bot in [dd7f76ac](https://github.com/getsentry/warden/commit/dd7f76ac25cd8d70f2557638bd8d8678d9776e96)
+- release: 0.22.0 by @getsentry-bot in [08a3b2f0](https://github.com/getsentry/warden/commit/08a3b2f0139248c80ba499038320556bedb66347)
+- release: 0.21.0 by @getsentry-bot in [2795626e](https://github.com/getsentry/warden/commit/2795626e8f1ff3c5a7a041d0ecb9ae3c047641cd)
+- Docs UI improvements by @PickleNik in [#248](https://github.com/getsentry/warden/pull/248)
+- release: 0.20.0 by @getsentry-bot in [e7dbd6eb](https://github.com/getsentry/warden/commit/e7dbd6eb1680d1d15df6388084237e5d231ec929)
+- release: 0.19.0 by @getsentry-bot in [0096b702](https://github.com/getsentry/warden/commit/0096b7024749edf9e62a791340cfea2ef5b899bf)
+- chores(docs): Add instruction for creating org rulesets by @Jeffreyhung in [#244](https://github.com/getsentry/warden/pull/244)
+- release: 0.18.0 by @getsentry-bot in [e3691621](https://github.com/getsentry/warden/commit/e3691621d9f92de1f47fadb47be162b90f7042ff)
+- release: 0.17.0 by @getsentry-bot in [d6132a96](https://github.com/getsentry/warden/commit/d6132a960cb033ec737042bea68f195b5cadef76)
+- release: 0.16.0 by @getsentry-bot in [62909fe4](https://github.com/getsentry/warden/commit/62909fe4d28c322508945382d001d0a6844852f2)
+- release: 0.15.0 by @getsentry-bot in [d9905cee](https://github.com/getsentry/warden/commit/d9905ceee868cba2aa41f8f17233eacb44901c36)
+- release: 0.14.0 by @getsentry-bot in [538f13b4](https://github.com/getsentry/warden/commit/538f13b44a4e78f30ecc668957f6cede7d8e7161)
+- release: 0.13.0 by @getsentry-bot in [063517a9](https://github.com/getsentry/warden/commit/063517a90a903b56919eec66542b129950ff866c)
+- release: 0.12.0 by @getsentry-bot in [7177c99f](https://github.com/getsentry/warden/commit/7177c99fb10de1cbec8b788881d501a4f8d72965)
+- Remove outdated SPEC.md by @dcramer in [#181](https://github.com/getsentry/warden/pull/181)
+- release: 0.11.0 by @getsentry-bot in [b5525e63](https://github.com/getsentry/warden/commit/b5525e6332a7090940f066b676ea4cf587aa1e2a)
+- release: 0.10.0 by @getsentry-bot in [5762bc7a](https://github.com/getsentry/warden/commit/5762bc7a0b608cdbe31db7b740e8eeff21da2081)
+- Add repository and source context to skill metrics by @dcramer in [#165](https://github.com/getsentry/warden/pull/165)
+
+_Plus 5 more_
+
 ## 0.26.2
 
 ### New Features ✨

package/TELEMETRY.md

@@ -0,0 +1,323 @@
+---
+spec: https://github.com/getsentry/junior/blob/main/TELEMETRY.spec.md
+---
+
+# Telemetry
+
+## Goal
+
+Use this when investigating Warden production incidents in the CLI or GitHub
+Action. Start with a Sentry event, trace ID, repository, GitHub Action run,
+skill or trigger name, file path, finding ID, or model/tool symptom, then use
+the recipes below to find the failing run and next query.
+
+Primary backend: Sentry Logs, Issues, Spans/Traces, and Metrics. Local CLI runs
+also write `.warden/logs/*.jsonl`; those logs can carry the Sentry `traceId`
+when telemetry is enabled.
+
+## Where To Query
+
+| Starting Point | Query Surface | Pivot | Answers | Next Step |
+| -------------- | ------------- | ----- | ------- | --------- |
+| `trace_id` from CLI summary, JSONL, or `Workflow initialized` | Sentry Traces and Logs | `span_id` | full run timeline, slow/error span | inspect skill or workflow span |
+| Sentry `event_id` | Sentry Issues/Event | `trace_id`, `operation`, `warden.trigger.name`, `gen_ai.agent.name` | exception context and owning workflow | query trace logs |
+| GitHub repository or Action run | Sentry Logs and Metrics | `vcs.owner.name`, `vcs.repository.name`, `cicd.pipeline.run.id`, `github.event.name` | recent Warden runs and trigger count | open matching trace |
+| Skill or trigger name | Sentry Spans, Issues, Metrics | `gen_ai.agent.name`, `warden.trigger.name` | failing skill, model cost, finding count | inspect hunk or agent spans |
+| File path or hunk | Sentry Spans | `code.file.path`, `warden.hunk.line_range` | hunk analysis state and extraction failures | inspect agent span |
+| Model, tool, or token symptom | Sentry Spans | `gen_ai.*`, `gen_ai.tool.name` | Claude turn, tool, cost, and token behavior | inspect child spans |
+| Stale comment or fix-eval symptom | Sentry Spans and Metrics | `warden.fix_eval.finding_id`, `warden.fix_eval.verdict` | whether a finding was evaluated or resolved | inspect fix eval span |
+
+## Investigation Pivots
+
+| Pivot | Meaning | Found In | First Query |
+| ----- | ------- | -------- | ----------- |
+| `trace_id` | one Warden run trace | CLI verbose summary, JSONL, logs, issues, spans | open trace |
+| `span_id` | one workflow, skill, hunk, model, or tool span | logs, spans | inspect span |
+| `event_id` | captured Sentry error | Sentry issue/event | open event |
+| `vcs.repository.name` | repository name | global attributes, logs, metrics, spans | repo runs |
+| `vcs.owner.name` | repository owner or org | global attributes, logs, metrics, spans | repo runs |
+| `vcs.repository.url.full` | canonical repository URL | global attributes, logs, metrics, spans | exact repo runs |
+| `github.event.name` | Action event name | `workflow.run` span | action entry |
+| `cicd.pipeline.run.id` | GitHub Actions run ID | global attributes, logs, metrics, spans | action run |
+| `cicd.pipeline.run.url.full` | GitHub Actions run URL | global attributes, logs, metrics, spans | action run |
+| `warden.trigger.name` | matched Warden trigger | trigger exceptions | trigger failures |
+| `gen_ai.agent.name` | configured or resolved skill/agent | spans, issues, metrics | skill timeline |
+| `code.file.path` | file being analyzed or judged | hunk/file/fix spans | file analysis |
+| `gen_ai.conversation.id` | Claude Code SDK session ID | `gen_ai.invoke_agent` span | agent session |
+| `warden.fix_eval.finding_id` | finding comment identity | fix evaluation spans | fix verdict |
+
+## Query Recipes
+
+Trace log history after opening a Sentry event, CLI JSONL run, or verbose CLI
+summary.
+
+```text
+dataset=logs query='trace_id:"<trace_id>"'
+fields=timestamp,level,message,trace_id,span_id,vcs.owner.name,vcs.repository.name,gen_ai.agent.name,warden.trigger.name,error.type,exception.message
+sort=timestamp
+```
+
+Recent GitHub Action runs for a repository.
+
+```text
+dataset=logs query='message:"Workflow initialized" vcs.owner.name:"<owner>" vcs.repository.name:"<repo>"'
+fields=timestamp,trace_id,github.event.name,cicd.pipeline.run.id,cicd.pipeline.run.url.full,warden.trigger.count,release,environment
+sort=-timestamp
+```
+
+Workflow run spans for a repository.
+
+```text
+dataset=spans query='span.op:workflow.run vcs.owner.name:"<owner>" vcs.repository.name:"<repo>"'
+fields=timestamp,trace,span_id,span.duration,github.event.name,cicd.pipeline.run.id,cicd.pipeline.run.url.full,warden.trigger.count,warden.finding.count,error.type
+sort=-timestamp
+```
+
+Skill execution timeline for a slow or failing skill.
+
+```text
+dataset=spans query='span.op:skill.run gen_ai.agent.name:"<skill_name>"'
+fields=timestamp,trace,span_id,span.duration,gen_ai.agent.name,warden.trigger.name,warden.file.count,warden.finding.count,error.type
+sort=-timestamp
+```
+
+`warden.trigger.name` is present only for trigger-backed runs. Direct CLI skill
+runs have `gen_ai.agent.name` without trigger metadata.
+
+File or hunk analysis for a suspicious path.
+
+```text
+dataset=spans query='span.op:skill.analyze_hunk code.file.path:"<path>"'
+fields=timestamp,trace,span_id,span.duration,gen_ai.agent.name,warden.hunk.line_range,warden.hunk.failed,warden.finding.count,error.type
+sort=-timestamp
+```
+
+Agent/model calls for token, cost, or provider symptoms.
+
+```text
+dataset=spans query='span.op:gen_ai.invoke_agent gen_ai.agent.name:"<skill_name>"'
+fields=timestamp,trace,span_id,span.duration,gen_ai.conversation.id,gen_ai.request.model,gen_ai.response.model,gen_ai.usage.total_tokens,gen_ai.cost.total_tokens,error.type
+sort=-timestamp
+```
+
+Structured auxiliary calls use `span.op:gen_ai.chat` and include
+`warden.ai.task` (`extraction`, `deduplication`, `fix_quality`,
+`fix_evaluation`, `consolidation`, or `skill_build`) when available.
+
+Tool calls inside a Claude Code SDK turn or structured tool loop.
+
+```text
+dataset=spans query='span.op:gen_ai.execute_tool gen_ai.tool.name:"<tool_name>"'
+fields=timestamp,trace,span_id,span.duration,gen_ai.agent.name,warden.ai.task,gen_ai.tool.name,tool.elapsed_seconds,error.type
+sort=-timestamp
+```
+
+Captured trigger or workflow exceptions.
+
+```text
+dataset=issues query='warden.trigger.name:"<trigger_name>" OR gen_ai.agent.name:"<skill_name>" OR operation:"<operation>"'
+fields=timestamp,event_id,trace_id,operation,warden.trigger.name,gen_ai.agent.name,error.type,exception.message
+sort=-timestamp
+```
+
+Finding fix evaluation and stale comment resolution.
+
+```text
+dataset=spans query='span.op:fix_eval.evaluate warden.fix_eval.finding_id:"<finding_id>"'
+fields=timestamp,trace,span_id,span.duration,code.file.path,code.line.number,gen_ai.agent.name,warden.fix_eval.verdict,warden.fix_eval.used_fallback,error.type
+sort=-timestamp
+```
+
+Repository-level health and cost.
+
+```text
+dataset=metrics query='metric:warden.workflow.runs OR metric:warden.skill.duration OR metric:warden.gen_ai.cost.usd vcs.owner.name:"<owner>" vcs.repository.name:"<repo>"'
+fields=timestamp,metric,vcs.owner.name,vcs.repository.name,cicd.pipeline.run.id,gen_ai.agent.name,gen_ai.request.model,value
+sort=-timestamp
+```
+
+Total findings in a time window, segmented by skill and repository. Use the
+Sentry time picker for the window. Query `skill.run` spans so the count uses
+the final post-processed findings, not per-hunk candidates.
+
+```text
+dataset=spans query='span.op:skill.run warden.finding.count:>0'
+fields=timestamp,trace,span_id,span.duration,vcs.owner.name,vcs.repository.name,gen_ai.agent.name,warden.trigger.name,warden.finding.count
+aggregate=sum(warden.finding.count) by gen_ai.agent.name,vcs.owner.name,vcs.repository.name
+sort=-timestamp
+```
+
+Add `gen_ai.agent.name:"<skill_name>"`, `vcs.owner.name:"<owner>"`, or
+`vcs.repository.name:"<repo>"` to narrow the same query.
+
+Fix evaluation verdict breakdown for a skill. Use the Sentry time picker for
+the window.
+
+```text
+dataset=spans query='span.op:fix_eval.evaluate gen_ai.agent.name:"<skill_name>"'
+fields=timestamp,trace,span_id,span.duration,vcs.owner.name,vcs.repository.name,gen_ai.agent.name,code.file.path,code.line.number,warden.fix_eval.finding_id,warden.fix_eval.verdict,warden.fix_eval.used_fallback
+aggregate=count() by warden.fix_eval.verdict,gen_ai.agent.name,vcs.owner.name,vcs.repository.name
+sort=-timestamp
+```
+
+Finding lifecycle from analysis to fix evaluation. First find the evaluation
+span by finding ID, then open the trace or query the trace ID with the path and
+skill from that span.
+
+```text
+dataset=spans query='span.op:fix_eval.evaluate warden.fix_eval.finding_id:"<finding_id>"'
+fields=timestamp,trace,span_id,span.duration,vcs.owner.name,vcs.repository.name,gen_ai.agent.name,code.file.path,code.line.number,warden.fix_eval.finding_id,warden.fix_eval.verdict
+sort=-timestamp
+```
+
+```text
+dataset=spans query='trace:"<trace_id>" (span.op:skill.run OR span.op:skill.analyze_hunk OR span.op:fix_eval.evaluate) gen_ai.agent.name:"<skill_name>"'
+fields=timestamp,span.op,span_id,span.duration,code.file.path,warden.hunk.line_range,warden.finding.count,warden.fix_eval.finding_id,warden.fix_eval.verdict,error.type
+sort=timestamp
+```
+
+## Domains
+
+### Workflow Entry
+
+The CLI or GitHub Action did not start, selected no work, or failed while
+building repository context.
+
+Events: `Workflow initialized`, top-level CLI/action fatal error
+
+Spans: `workflow.run`, `workflow.init`, `config.load`
+
+Attributes: `trace_id`, `vcs.owner.name`, `vcs.repository.name`,
+`vcs.repository.url.full`, `warden.source`, `github.event.name`,
+`cicd.pipeline.name`, `cicd.pipeline.run.id`,
+`cicd.pipeline.run.url.full`, `warden.trigger.count`,
+`warden.finding.count`
+
+### Trigger And GitHub Review
+
+The Action ran, but checks, comments, review posting, or trigger execution
+failed.
+
+Events: operation tags `create_core_check`, `fetch_existing_comments`,
+`post_thread_reply`, `dismiss_review`, `update_core_check`
+
+Spans: `workflow.setup`, `workflow.execute`, `trigger.execute`,
+`workflow.review`
+
+Attributes: `warden.trigger.name`, `gen_ai.agent.name`, `operation`,
+`vcs.owner.name`, `vcs.repository.name`
+
+### Skill Analysis
+
+A skill was slow, returned no findings, failed every hunk, or analyzed the
+wrong files.
+
+Events: `Skill execution started`, `Skill execution complete`
+
+Spans: `skill.run`, `skill.analyze_file`, `skill.analyze_hunk`
+
+Attributes: `gen_ai.agent.name`, `warden.file.count`, `code.file.path`,
+`warden.hunk.count`, `warden.hunk.line_range`, `warden.hunk.failed`,
+`warden.finding.count`
+
+### Agent And Model
+
+Claude Code SDK execution, Anthropic calls, model choice, tokens, tool use, or
+provider failures look wrong.
+
+Events: SDK/runtime errors captured on the owning skill or trigger
+
+Spans: `gen_ai.invoke_agent`, `gen_ai.chat`, `gen_ai.execute_tool`
+
+Attributes: `gen_ai.agent.name`, `gen_ai.conversation.id`,
+`gen_ai.request.model`, `gen_ai.response.model`, `gen_ai.usage.input_tokens`,
+`gen_ai.usage.output_tokens`, `gen_ai.usage.total_tokens`,
+`gen_ai.tool.name`, `tool.elapsed_seconds`
+
+### Finding Pipeline
+
+Findings were extracted, deduplicated, verified, merged, or stripped
+unexpectedly.
+
+Events: `Suggested fix quality gate`
+
+Spans: skill spans plus auxiliary `gen_ai.invoke_agent` spans
+
+Metrics: `warden.extraction.attempts`, `warden.extraction.findings`,
+`warden.dedup.total`, `warden.dedup.unique`, `warden.dedup.removed`,
+`warden.fix_gate.checked`, `warden.fix_gate.stripped_deterministic`,
+`warden.fix_gate.stripped_semantic`, `warden.fix_gate.semantic_unavailable`
+
+Attributes: `gen_ai.agent.name`, `warden.extraction.method`,
+`warden.fix_gate.checked`, `warden.fix_gate.stripped_deterministic`,
+`warden.fix_gate.stripped_semantic`, `warden.fix_gate.semantic_unavailable`
+
+### Fix Evaluation And Stale Comments
+
+Existing Warden comments were not resolved, were judged incorrectly, or fix
+evaluation failed.
+
+Events: operation tags `fetch_fix_context`, `evaluate_fix_attempts`,
+`resolve_stale_comments`
+
+Spans: `workflow.resolve`, `fix_eval.run`, `fix_eval.evaluate`
+
+Attributes: `warden.fix_eval.comment_count`, `warden.fix_eval.finding_id`,
+`gen_ai.agent.name`, `warden.fix_eval.verdict`,
+`warden.fix_eval.used_fallback`, `code.file.path`, `code.line.number`
+
+### Local Run Logs
+
+A local CLI report exists, but the matching Sentry trace or run metadata is
+needed.
+
+Events: JSONL records in `.warden/logs/*.jsonl`
+
+Spans: `skill.run`, `skill.analyze_file`, `skill.analyze_hunk`,
+`gen_ai.invoke_agent`
+
+Attributes: `traceId` in JSONL, `runId`, `headSha`, `model`,
+`gen_ai.agent.name` in telemetry
+
+## Configuration
+
+| Setting | Controls | Default |
+| ------- | -------- | ------- |
+| `WARDEN_SENTRY_DSN` | Enables Sentry logs, issues, traces, and metrics | disabled |
+| `WARDEN_MODEL` | Fallback model recorded on gen AI spans and JSONL | SDK default when unset |
+| `WARDEN_ANTHROPIC_API_KEY` | Anthropic auth for CI and auxiliary calls | falls back to `ANTHROPIC_API_KEY` or Claude auth |
+| `ANTHROPIC_API_KEY` | Secondary Anthropic auth source | unset |
+| `GITHUB_REPOSITORY` | Action repository scope for `vcs.*` attributes | GitHub Actions only |
+| `GITHUB_EVENT_NAME` | Action event and `github.event.name` attribute | GitHub Actions only |
+| `GITHUB_RUN_ID` | Action run scope for `cicd.pipeline.run.*` attributes | GitHub Actions only |
+| `GITHUB_WORKFLOW` | Action workflow name for `cicd.pipeline.name` | GitHub Actions only |
+| `GITHUB_JOB` | Action job name for `cicd.pipeline.task.name` | GitHub Actions only |
+| CLI `--output` | Explicit JSONL output location | `.warden/logs/` run file |
+
+## Attribute Notes
+
+- `vcs.*`, `code.*`, and `gen_ai.*` fields follow OpenTelemetry semantic
+  conventions where applicable.
+- `cicd.*` fields follow OpenTelemetry CI/CD semantic conventions for GitHub
+  Actions run metadata where there is a direct match.
+- `warden.*` fields are Warden-owned local attributes for concepts that do not
+  have OpenTelemetry semantic attributes, such as triggers, hunks, finding
+  counts, and fix evaluation.
+- `github.event.name` is GitHub-specific because OpenTelemetry does not define
+  the workflow trigger event name as a standard CI/CD attribute.
+- `gen_ai.request.messages` and `gen_ai.response.text` may contain prompt or
+  model text. Use IDs, models, tokens, and status fields for triage unless the
+  incident specifically requires content inspection.
+- `gen_ai.cost.total_tokens` stores SDK-reported USD cost despite the inherited
+  attribute name.
+- `traceId` in JSONL is the same production pivot as Sentry `trace_id`.
+
+## References
+
+- [OpenTelemetry GenAI semantic conventions](https://opentelemetry.io/docs/specs/semconv/gen-ai/)
+- [OpenTelemetry GenAI agent spans](https://opentelemetry.io/docs/specs/semconv/gen-ai/gen-ai-agent-spans/)
+- [OpenTelemetry VCS attributes](https://opentelemetry.io/docs/specs/semconv/registry/attributes/vcs/)
+- [OpenTelemetry code attributes](https://opentelemetry.io/docs/specs/semconv/registry/attributes/code/)
+- [OpenTelemetry CI/CD attributes](https://opentelemetry.io/docs/specs/semconv/registry/attributes/cicd/)
+- [Sentry AI Agents module](https://develop.sentry.dev/sdk/telemetry/traces/modules/ai-agents/)
+- [Sentry JavaScript Node SDK logs](https://docs.sentry.io/platforms/javascript/guides/node/logs/)

package/dist/cli/main.d.ts

@@ -21,6 +21,7 @@ export declare function resolveInvocationCwd(baseCwd: string, cliCwd: string | u
 export interface RunSkillSpec {
     name: string;
     displayName?: string;
+    triggerName?: string;
     skill: string;
     remote?: string;
     failOn?: SeverityThreshold;

package/dist/cli/main.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"main.d.ts","sourceRoot":"","sources":["../../src/cli/main.ts"],"names":[],"mappings":"AAKA,OAAO,KAAK,EAAE,eAAe,EAAE,YAAY,EAAE,MAAM,qBAAqB,CAAC;AACzE,OAAO,EAA8C,KAAK,kBAAkB,EAA4B,MAAM,kBAAkB,CAAC;AAKjI,OAAO,KAAK,EAAE,WAAW,EAAE,iBAAiB,EAAE,mBAAmB,EAAuB,MAAM,mBAAmB,CAAC;AAIlH,OAAO,EAA8C,KAAK,UAAU,EAAE,MAAM,WAAW,CAAC;AAExF,OAAO,EAAE,sBAAsB,EAAyB,MAAM,cAAc,CAAC;AAG7E,OAAO,EACL,QAAQ,EAIR,aAAa,EAcb,KAAK,gBAAgB,EACtB,MAAM,mBAAmB,CAAC;AAoB3B;;;GAGG;AACH,eAAO,MAAM,eAAe,iBAAwB,CAAC;AAErD;;;GAGG;AACH,eAAO,MAAM,WAAW;;CAAmB,CAAC;AAiC5C,wEAAwE;AACxE,wBAAgB,oBAAoB,CAAC,OAAO,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,GAAG,SAAS,GAAG,MAAM,CAExF;AAiYD,MAAM,WAAW,YAAY;IAC3B,IAAI,EAAE,MAAM,CAAC;IACb,WAAW,CAAC,EAAE,MAAM,CAAC;IACrB,KAAK,EAAE,MAAM,CAAC;IACd,MAAM,CAAC,EAAE,MAAM,CAAC;IAChB,MAAM,CAAC,EAAE,iBAAiB,CAAC;IAC3B,aAAa,CAAC,EAAE,mBAAmB,CAAC;IACpC,OAAO,EAAE,OAAO,CAAC,UAAU,CAAC,OAAO,sBAAsB,CAAC,CAAC,CAAC;IAC5D,aAAa,EAAE,kBAAkB,CAAC;CACnC;AAED,UAAU,gBAAgB;IACxB,OAAO,EAAE,WAAW,EAAE,CAAC;IACvB,eAAe,EAAE,WAAW,EAAE,CAAC;IAC/B,UAAU,EAAE,OAAO,CAAC;IACpB,cAAc,EAAE,MAAM,EAAE,CAAC;CAC1B;AAED,KAAK,0BAA0B,GAAG,IAAI,CACpC,kBAAkB,EAClB,OAAO,GAAG,UAAU,GAAG,SAAS,GAAG,gBAAgB,GAAG,gBAAgB,GAAG,qBAAqB,GAAG,gBAAgB,CAClH,CAAC;AA0BF,yDAAyD;AACzD,wBAAgB,iBAAiB,CAC/B,KAAK,EAAE,IAAI,CAAC,eAAe,EAAE,SAAS,CAAC,EACvC,QAAQ,CAAC,EAAE,MAAM,GAChB,MAAM,GAAG,SAAS,CAepB;AAED,gFAAgF;AAChF,wBAAgB,uBAAuB,CACrC,IAAI,EAAE,kBAAkB,EACxB,SAAS,EAAE,0BAA0B,GACpC,kBAAkB,CAcpB;AAkED,2DAA2D;AAC3D,wBAAsB,gBAAgB,CAAC,IAAI,EAAE;IAC3C,KAAK,EAAE,YAAY,EAAE,CAAC;IACtB,QAAQ,CAAC,EAAE,MAAM,CAAC;IAClB,OAAO,EAAE,UAAU,CAAC;IACpB,QAAQ,EAAE,MAAM,CAAC;IACjB,QAAQ,EAAE,QAAQ,CAAC;CACpB,GAAG,OAAO,CAAC,gBAAgB,EAAE,CAAC,CAa9B;AAED,qFAAqF;AACrF,wBAAgB,sBAAsB,CACpC,MAAM,EAAE,IAAI,CAAC,YAAY,EAAE,UAAU,CAAC,GAAG,IAAI,GAAG,SAAS,EACzD,QAAQ,CAAC,EAAE,MAAM,GAChB,MAAM,GAAG,SAAS,CAOpB;AAED,6EAA6E;AAC7E,wBAAgB,+BAA+B,CAC7C,MAAM,EAAE,IAAI,CAAC,YAAY,EAAE,UAAU,CAAC,GAAG,IAAI,GAAG,SAAS,GACxD,MAAM,GAAG,SAAS,CAEpB;AAED,0FAA0F;AAC1F,wBAAgB,+BAA+B,CAC7C,MAAM,EAAE,IAAI,CAAC,YAAY,EAAE,UAAU,CAAC,GAAG,IAAI,GAAG,SAAS,GACxD,MAAM,GAAG,SAAS,CAKpB;AAED,wFAAwF;AACxF,wBAAgB,kBAAkB,CAChC,MAAM,EAAE,IAAI,CAAC,YAAY,EAAE,UAAU,CAAC,GAAG,IAAI,GAAG,SAAS,EACzD,QAAQ,CAAC,EAAE,MAAM,GAChB,MAAM,CAER;AAED;;;GAGG;AACH,wBAAgB,kBAAkB,CAChC,OAAO,EAAE,OAAO,CAAC,UAAU,CAAC,OAAO,aAAa,CAAC,CAAC,EAClD,QAAQ,EAAE,UAAU,CAAC,UAAU,CAAC,EAChC,aAAa,CAAC,EAAE,mBAAmB,GAClC,gBAAgB,CA8BlB;AAuGD,4EAA4E;AAC5E,wBAAsB,SAAS,CAC7B,OAAO,EAAE,OAAO,CAAC,UAAU,CAAC,OAAO,sBAAsB,CAAC,CAAC,EAC3D,OAAO,EAAE,UAAU,EACnB,QAAQ,EAAE,QAAQ,GACjB,OAAO,CAAC,MAAM,CAAC,CAiNjB;AA8aD,oFAAoF;AACpF,wBAAsB,IAAI,IAAI,OAAO,CAAC,IAAI,CAAC,CA2G1C"}
+{"version":3,"file":"main.d.ts","sourceRoot":"","sources":["../../src/cli/main.ts"],"names":[],"mappings":"AAKA,OAAO,KAAK,EAAE,eAAe,EAAE,YAAY,EAAE,MAAM,qBAAqB,CAAC;AACzE,OAAO,EAA8C,KAAK,kBAAkB,EAA4B,MAAM,kBAAkB,CAAC;AAKjI,OAAO,KAAK,EAAE,WAAW,EAAE,iBAAiB,EAAE,mBAAmB,EAAuB,MAAM,mBAAmB,CAAC;AAIlH,OAAO,EAA8C,KAAK,UAAU,EAAE,MAAM,WAAW,CAAC;AAExF,OAAO,EAAE,sBAAsB,EAAyB,MAAM,cAAc,CAAC;AAG7E,OAAO,EACL,QAAQ,EAIR,aAAa,EAcb,KAAK,gBAAgB,EACtB,MAAM,mBAAmB,CAAC;AAoB3B;;;GAGG;AACH,eAAO,MAAM,eAAe,iBAAwB,CAAC;AAErD;;;GAGG;AACH,eAAO,MAAM,WAAW;;CAAmB,CAAC;AAiC5C,wEAAwE;AACxE,wBAAgB,oBAAoB,CAAC,OAAO,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,GAAG,SAAS,GAAG,MAAM,CAExF;AAiYD,MAAM,WAAW,YAAY;IAC3B,IAAI,EAAE,MAAM,CAAC;IACb,WAAW,CAAC,EAAE,MAAM,CAAC;IACrB,WAAW,CAAC,EAAE,MAAM,CAAC;IACrB,KAAK,EAAE,MAAM,CAAC;IACd,MAAM,CAAC,EAAE,MAAM,CAAC;IAChB,MAAM,CAAC,EAAE,iBAAiB,CAAC;IAC3B,aAAa,CAAC,EAAE,mBAAmB,CAAC;IACpC,OAAO,EAAE,OAAO,CAAC,UAAU,CAAC,OAAO,sBAAsB,CAAC,CAAC,CAAC;IAC5D,aAAa,EAAE,kBAAkB,CAAC;CACnC;AAED,UAAU,gBAAgB;IACxB,OAAO,EAAE,WAAW,EAAE,CAAC;IACvB,eAAe,EAAE,WAAW,EAAE,CAAC;IAC/B,UAAU,EAAE,OAAO,CAAC;IACpB,cAAc,EAAE,MAAM,EAAE,CAAC;CAC1B;AAED,KAAK,0BAA0B,GAAG,IAAI,CACpC,kBAAkB,EAClB,OAAO,GAAG,UAAU,GAAG,SAAS,GAAG,gBAAgB,GAAG,gBAAgB,GAAG,qBAAqB,GAAG,gBAAgB,CAClH,CAAC;AA0BF,yDAAyD;AACzD,wBAAgB,iBAAiB,CAC/B,KAAK,EAAE,IAAI,CAAC,eAAe,EAAE,SAAS,CAAC,EACvC,QAAQ,CAAC,EAAE,MAAM,GAChB,MAAM,GAAG,SAAS,CAepB;AAED,gFAAgF;AAChF,wBAAgB,uBAAuB,CACrC,IAAI,EAAE,kBAAkB,EACxB,SAAS,EAAE,0BAA0B,GACpC,kBAAkB,CAcpB;AAmED,2DAA2D;AAC3D,wBAAsB,gBAAgB,CAAC,IAAI,EAAE;IAC3C,KAAK,EAAE,YAAY,EAAE,CAAC;IACtB,QAAQ,CAAC,EAAE,MAAM,CAAC;IAClB,OAAO,EAAE,UAAU,CAAC;IACpB,QAAQ,EAAE,MAAM,CAAC;IACjB,QAAQ,EAAE,QAAQ,CAAC;CACpB,GAAG,OAAO,CAAC,gBAAgB,EAAE,CAAC,CAa9B;AAED,qFAAqF;AACrF,wBAAgB,sBAAsB,CACpC,MAAM,EAAE,IAAI,CAAC,YAAY,EAAE,UAAU,CAAC,GAAG,IAAI,GAAG,SAAS,EACzD,QAAQ,CAAC,EAAE,MAAM,GAChB,MAAM,GAAG,SAAS,CAOpB;AAED,6EAA6E;AAC7E,wBAAgB,+BAA+B,CAC7C,MAAM,EAAE,IAAI,CAAC,YAAY,EAAE,UAAU,CAAC,GAAG,IAAI,GAAG,SAAS,GACxD,MAAM,GAAG,SAAS,CAEpB;AAED,0FAA0F;AAC1F,wBAAgB,+BAA+B,CAC7C,MAAM,EAAE,IAAI,CAAC,YAAY,EAAE,UAAU,CAAC,GAAG,IAAI,GAAG,SAAS,GACxD,MAAM,GAAG,SAAS,CAKpB;AAED,wFAAwF;AACxF,wBAAgB,kBAAkB,CAChC,MAAM,EAAE,IAAI,CAAC,YAAY,EAAE,UAAU,CAAC,GAAG,IAAI,GAAG,SAAS,EACzD,QAAQ,CAAC,EAAE,MAAM,GAChB,MAAM,CAER;AAED;;;GAGG;AACH,wBAAgB,kBAAkB,CAChC,OAAO,EAAE,OAAO,CAAC,UAAU,CAAC,OAAO,aAAa,CAAC,CAAC,EAClD,QAAQ,EAAE,UAAU,CAAC,UAAU,CAAC,EAChC,aAAa,CAAC,EAAE,mBAAmB,GAClC,gBAAgB,CA8BlB;AAuGD,4EAA4E;AAC5E,wBAAsB,SAAS,CAC7B,OAAO,EAAE,OAAO,CAAC,UAAU,CAAC,OAAO,sBAAsB,CAAC,CAAC,EAC3D,OAAO,EAAE,UAAU,EACnB,QAAQ,EAAE,QAAQ,GACjB,OAAO,CAAC,MAAM,CAAC,CAiNjB;AA+aD,oFAAoF;AACpF,wBAAsB,IAAI,IAAI,OAAO,CAAC,IAAI,CAAC,CA2G1C"}

package/dist/cli/main.js

@@ -1,7 +1,7 @@
 import { existsSync, readFileSync, renameSync, unlinkSync, writeFileSync } from 'node:fs';
 import { dirname, join, relative, resolve } from 'node:path';
 import { config as dotenvConfig } from 'dotenv';
-import { Sentry, flushSentry, setGlobalAttributes, emitRunMetric, getTraceId } from '../sentry.js';
+import { Sentry, flushSentry, setRepositoryScope, emitRunMetric, getTraceId } from '../sentry.js';
 import { emptyToUndefined, loadWardenConfig, resolveSkillConfigs } from '../config/loader.js';
 import { verifyAuth } from '../sdk/runner.js';
 import { mapExtractionErrorCode } from '../sdk/errors.js';
@@ -479,6 +479,7 @@ async function createDirectSkillTask(args) {
     return {
         name: spec.name,
         displayName: spec.displayName,
+        triggerName: spec.triggerName,
         failOn: spec.failOn,
         minConfidence: spec.minConfidence,
         resolveSkill: async () => skill,
@@ -745,7 +746,7 @@ export async function runSkills(context, options, reporter) {
         skillsToRun = [];
     }
     // Set global telemetry context and emit run metric
-    setGlobalAttributes({ 'warden.repository': context.repository.fullName });
+    setRepositoryScope(context.repository.fullName);
     emitRunMetric();
     // Handle case where no skills to run
     if (skillsToRun.length === 0) {
@@ -993,7 +994,7 @@ async function runConfigMode(options, reporter) {
     }
     reporter.contextFiles(pullRequest.files);
     // Set global telemetry context and emit run metric
-    setGlobalAttributes({ 'warden.repository': context.repository.fullName });
+    setRepositoryScope(context.repository.fullName);
     emitRunMetric();
     // Resolve skills into triggers and match
     const resolvedTriggers = resolveSkillConfigs(config, options.model);
@@ -1049,6 +1050,7 @@ async function runConfigMode(options, reporter) {
     const specs = triggersToRun.map((trigger) => ({
         name: trigger.name,
         displayName: trigger.skill,
+        triggerName: trigger.name,
         skill: trigger.skill,
         remote: trigger.remote,
         failOn: trigger.failOn ?? options.failOn,