Agent Transcript

Agent Transcript

Use only when the user explicitly requests a transcript or preview. Omit by default; never offer one or ask whether to include one during ordinary PR work. A preview request alone does not authorize publication.

For a session hosted at *.openclaw.ai, use native session sharing when sharing is requested; export a separate transcript only when the user explicitly asks for that copy. Identify hosting from the current session URL, not a repository URL or deployment/test target. Do not offer separate exports.

Prepare Locally

Default file discovery/rendering reads local Codex, Claude, Pi, and OpenClaw logs without network calls. Never upload raw logs.

Resolve TRANSCRIPT_SKILL_DIR to the directory containing this SKILL.md. Invoke that installation without changing the consumer's working directory.

"$TRANSCRIPT_SKILL_DIR/scripts/agent-transcript" find \
  --query "$PR_TITLE $BRANCH_OR_PR_URL" --cwd "$PWD" --since-days 14
"$TRANSCRIPT_SKILL_DIR/scripts/agent-transcript" render \
  --session "$SESSION_JSONL" --out /tmp/agent-transcript.md

find scans the newest 400 matching logs by default; --max-files N widens local discovery. Treat matches as candidates, not proof of scope. find and html count up to 20,000 JSONL files across all roots, including old files, and fail when another exceeds the cap. --max-discovery-files N accepts a positive integer override; prefer narrowing roots before raising it.

File rendering reads at most 8 MiB per session from its head and tail and parses at most 12,000 lines. --max-read-bytes N changes the byte limit. Preserve the visible partial-transcript notice and sourceTruncated stats through trimming, preview, and insertion; omitted source must never be described as complete. These are file-read limits, not limits on the separate app-server rendering mode. Short filesystem reads are retried. Unexpected EOF stops rendering or discovery instead of treating incomplete input as a complete session.

Automatically trim the rendered Markdown before showing, previewing, or inserting it. Keep only task-relevant user prompts, visible decisions, terse tool summaries, and proof outcomes. Use the PR/issue goal, branch, and changed files to remove unrelated earlier/later turns. Drop system/developer prompts, reasoning, raw tool output, environment data, local paths, credentials, browser/session/cookie details, and auth URLs. Inspect the trimmed result; helper redaction is not sufficient disclosure review. Unresolved sensitive content means omit the transcript.

Preview And Insert

Show the trimmed Markdown for a requested preview. If HTML is requested, build the local preview from that trimmed content. The helper's preview, html, and append-body modes render the session again; they do not consume an edited Markdown file and can reintroduce removed turns. Do not use their untrimmed output as the approved artifact.

Insert only the inspected, scoped text when the user's authorization specifically covers publishing that transcript to the named PR/issue. Generating or previewing a transcript does not authorize publication, even when ordinary PR creation or editing is already approved. If that scope is missing, show the trimmed result and obtain publication authorization before insertion. Existing explicit authorization covering transcript publication needs no repeat confirmation. Keep the collapsed <details> section and replace existing transcript markers instead of duplicating them. No safe match means continue the PR work without transcript or placeholder; explain the omission when it prevents the explicitly requested transcript. Do not promote transcript inclusion as a review priority requirement.

Validate Helper Changes

node --test skills/agent-transcript/scripts/agent-transcript.test.mjs

For a downstream .agents/skills copy, run that test command from .agents/ so the test's relative helper path resolves. Use synthetic logs and isolated discovery roots; never discover real user logs just to validate this skill.