observe-agent-fuse

name	observe-agent-fuse
description	Run MAPS Phase 7 Observe with Langfuse. Use when Codex needs to apply the base observe-agent workflow through Langfuse traces, sessions, scores, datasets, prompt versions, metrics, feedback, or self-hostable/open-source LLM observability.

Observe Agent+Fuse

Versioning

Current version: 0.1.0.

Follow semantic versioning for this skill:

• Patch: wording, examples, references, or small workflow clarifications.
• Minor: new outputs, new required steps, new helper behavior, or expanded workflow capability.
• Major: renamed outputs, changed artifact contracts, removed behavior, or incompatible workflow changes.

When changing this skill, update Current version and add a Changelog entry with the date, version, and short summary of behavior changed.

Changelog

• 2026-06-19 - v0.1.0 - Established the initial MAPS wrapper skill version baseline and changelog tracking.

Use /observe-agent+fuse after Phase 6 when the deployed agent should be observed through Langfuse.

This wrapper keeps /observe-agent as the base workflow and adds Langfuse-specific tracing, feedback, prompt/version, score, and self-hosting considerations.

Project foundation updates

At the start of every project run, look for project-foundation.md. If it exists, read Persistent Memory Contract and use its configured notes, sources, memory, RAG, and sync rules as the project defaults. If .maps/foundation-preferences.json exists, use it as the structured preference source for automation.

When this wrapper creates durable knowledge, write it through the shared MAPS memory helper. The helper gives this skill its own named note under the configured notes root, mirrors that note into the configured RAG location when one exists, appends MAPS Skill Run Log, and records a RAG reindex manifest.

At the end of the run, call the helper after creating the primary output artifact:

ash python "$CODEX_HOME/skills/foundation/scripts/maps_memory.py" complete-run --project . --skill /observe-agent+fuse --phase Wrapper --output "<primary artifact path>" --summary-file "<primary artifact path>" --memory-updates "<notes, sources, memory, or RAG updates>"

If the helper is unavailable, manually append the timestamp, skill, phase, output path, memory updates, and short note to project-foundation.md, then update this skill's named note in /maps-runs/.

• All /observe-agent inputs
• Langfuse project, environment, trace/session links, API/key availability, prompt/version links, score definitions, dataset links, and deployment mode when available

Required interview

Before acting, ask for any missing answers required to choose the target artifact, wrapper path, runtime, provider, evidence source, or approval boundary.

Ask exactly one question at a time. Do not present the user with a multi-question form, checklist, or table to fill out. Ask the next most important missing question, wait for the answer, then continue.

1. Run the base /observe-agent workflow first.
2. Connect Phase 7 evidence to Langfuse:
   • traces and sessions
   • scores and feedback
   • prompt versions
   • datasets and experiments
   • cost, latency, and usage metrics
   • cloud or self-hosted deployment notes
3. Recommend the Langfuse observation plan before asking broad questions.
4. Record missing Langfuse setup when access, tracing, score definitions, prompt tracking, feedback capture, or retention choices are unavailable.
5. Use Langfuse evidence to identify production failures, prompt regressions, user feedback patterns, cost/latency issues, and improvement candidates.
6. Convert confirmed findings into Phase 8 improvement backlog items.

Langfuse Must Cover

• Trace/session links for representative production requests.
• Score definitions and feedback capture.
• Prompt/version links when prompts are part of the agent behavior.
• Dataset or experiment links when production observations should become eval cases.
• Cloud versus self-hosted operating choice, including retention and ownership notes.
• Improvement items backed by Langfuse evidence.

Best Fit

• Choose this wrapper when developers should compare a self-hostable/open-source observability option.
• Choose it when prompt management, feedback, scoring, and data-control concerns matter.
• Do not assume Langfuse replaces LangSmith; make the tradeoff visible.

Completion report

When the skill is complete, tell the user explicitly. Do not end with only files changed or raw output.

Report:

• Completion status: complete, blocked, or needs more answers.
• Outcome: the concrete artifact, decision, scaffold, implementation, or plan produced.
• Key decisions or changes made.
• Memory update: whether the shared MAPS memory helper ran, what note/run log was updated, and what RAG or notes locations need syncing.
• Next skill: the next MAPS phase skill, wrapper, or follow-up skill that should run.

If the skill is blocked, say what answer, artifact, access, approval, or tool is needed before the next skill can run.

Output

• agents/{agent-handle}/observation-plan.md
• agents/{agent-handle}/observation-log.md
• Langfuse evidence links and missing setup notes
• Phase 8 improvement backlog entries or handoff notes