ruflo-observability
Reviewed
at commit 13af5f1
Caution: Its skills pre-approve Bash and have Claude run the unpinned npm package @claude-flow/cli@latest through npx, so each new version runs without asking.
Structured logging, distributed tracing, and metrics — correlate agent swarm activity with application telemetry. As a mod (ADR-445): a tighten-only guard on…
0 upvotes
Log in to upvote or save- Made by
- ruvnet
- Pinned commit
-
13af5f1208960c98832d6d9096a8d7ec45bcdc3e - What we checked
-
Refuses memory writes (spans, logs, metric labels) that hold a key, token, password, US social security number or card number, and adds a /observe-mod command that answers locally. It writes a small status file in your project. Its skills and commands read and store telemetry records in ruflo's memory.
Review details
Static review of the source at the pinned commit; not run. Delta review of the changes since 87d8687; the rest carries over from the review at that commit. The new version changes no hook code: it adds a repository field to plugin.json, bumps the version, and rewords the observe-trace skill and the agent so Claude no longer calls agentdb_context-synthesize for trace spans (it is removed from the skill's allowed-tools) and writes the summary itself. The hook module registers session.start (reads the project root, registers /observe-mod, writes .claude-flow/observe-mod/status.json), tool.call (when the guard option is on, the default; screens inputs of memory_store, agentdb_pattern-store, agentdb_hierarchical-store and agentdb_batch with local regex and entropy rules and can only deny, never rewrite or approve) and command.run for observe-mod (status and scan, answered locally). The hooks make no network call, spawn no process and call no model; findings are rule names, never the matched text; the status write is wrapped in try/catch. The skills observe-metrics and observe-trace pre-approve Bash alongside the ruflo memory and agentdb tools and show a CLI alternative, npx @claude-flow/cli@latest memory search ..., which is unpinned; the agent prose also tells Claude to run npx @claude-flow/cli@latest for memory and neural training commands. The skills tell Claude to treat stored span and log text as data, not instructions. The @claude-flow/cli package and the ruflo memory tools are not part of this plugin and were not reviewed. scripts/smoke.sh is a test script not loaded by the hooks.
- What it uses
-
- Reads your prompts: no
- Sees tool calls: yes
- Blocks or changes tool calls: yes
- Reads or writes files: yes
- Runs programs on your machine: yes
- Calls the model (uses tokens): no
- Sends data over the network: no
- Stores data: yes
- Shows something on screen: yes
Structured logging, distributed tracing, and metrics — correlate agent swarm activity with application telemetry. As a mod (ADR-445): a tighten-only guard on its tools, /observe-mod, and a status file the console shows
Made this mod? Add a badge to your README
Paste this into your README. The badge says reviewed once the current version has been reviewed, and links readers to this page.
[](https://modsforclaude.com/mods/ruflo-observability/)