Stats
Actions
Tags
From beads
Logs and labels agent interactions (prompts, responses, tool calls) to an append-only JSONL file. Subcommands: record and label.
How this command is triggered — by the user, by Claude, or both
Slash command
/beads:audit record|labelbeads/commands/The summary Claude sees in its command listing — used to decide when to auto-load this command
Append-only audit logging for agent interactions (prompts, responses, tool calls) in `.beads/interactions.jsonl`. Each line is one event. Labeling is done by appending a new `"label"` event referencing a previous entry. ## Usage - **Record an interaction**: - `bd audit record --kind llm_call --model "claude-3-5-haiku" --prompt "..." --response "..."` - `bd audit record --kind tool_call --tool-name "go test" --exit-code 1 --error "..." --issue-id bd-42` - **Pipe JSON via stdin**: - `cat event.json | bd audit record` - **Label an entry**: - `bd audit label int-a1b2 --label good -...
Append-only audit logging for agent interactions (prompts, responses, tool calls) in .beads/interactions.jsonl.
Each line is one event. Labeling is done by appending a new "label" event referencing a previous entry.
Record an interaction:
bd audit record --kind llm_call --model "claude-3-5-haiku" --prompt "..." --response "..."bd audit record --kind tool_call --tool-name "go test" --exit-code 1 --error "..." --issue-id bd-42Pipe JSON via stdin:
cat event.json | bd audit recordLabel an entry:
bd audit label int-a1b2 --label good --reason "Worked perfectly"bd audit label int-a1b2 --label bad --reason "Hallucinated a file path"bd dolt push includes .beads/interactions.jsonl in the commit allowlist.npx claudepluginhub gastownhall/beads --plugin beads/ng-traceCaptures an agent run's tool calls, decisions, inputs, outputs, token use, and approval steps as a repeatable trace record linked to verification claims.
/assist-logsQueries context-inject hook logs via local API to view injected knowledge and milestones in Claude Code sessions. Supports --limit/-n and --search/-s filters; checks API health first.
/sessionlogCaptures current session context as structured markdown log with key decisions, commands, problems, resolutions, and files changed. Saves to ./SessionLogs/session-DDMMYY-HH:MM.md
/logInteractively logs AI experiments by gathering tool, prompt, summary, and rating details conversationally, then displays formatted terminal report with ID and stats.
/buidl-traceDisplays agent execution trace for the current loop session from trace.jsonl, rendering table of timestamps, events, agents, phases, cycles, details, plus summary of events, agents, and errors.