session-corpus-audit
Analyze session quality trends — identify high-churn patterns, report waste, flag sessions exceeding 500 tool calls
Best use case
session-corpus-audit is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Analyze session quality trends — identify high-churn patterns, report waste, flag sessions exceeding 500 tool calls
Teams using session-corpus-audit should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/session-corpus-audit/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How session-corpus-audit Compares
| Feature / Agent | session-corpus-audit | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Analyze session quality trends — identify high-churn patterns, report waste, flag sessions exceeding 500 tool calls
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Session Corpus Audit Analyze session signals to identify quality trends and waste patterns. ## Data source Session signals live at `.Codex/state/session-signals/YYYY-MM-DD.jsonl`. Each line is a JSON object with: session_id, transcript_path, cwd, permission_mode, hook_event_name, stop_hook_active, last_assistant_message. ## Audit procedure ### 1. Collect recent signals ```bash # Last 7 days of session signals for f in $(ls -t .Codex/state/session-signals/*.jsonl | head -7); do echo "=== $(basename $f) ===" wc -l "$f" cat "$f" done ``` ### 2. Identify high-churn sessions - Sessions that fired multiple Stop hooks (restarts/crashes) - Sessions with last_assistant_message indicating errors or blocks - Sessions ending with permission denials ### 3. Estimate tool-call volume - Check `.Codex/state/session-governor/tool-call-count` for daily totals - Flag any day exceeding 500 tool calls (potential runaway session) ### 4. Detect recurring patterns - Same error messages across sessions (systemic issues) - Sessions that ended mid-task (unreleased wip labels, uncommitted changes) - Permission mode patterns (bypassPermissions vs default) ### 5. Produce quality report Output a markdown report with: - Session count by day (last 7 days) - High-churn sessions with root cause - Recurring error patterns - Waste estimate (sessions that produced no commits) - Recommendations for workflow improvement ## When to use - Weekly quality review - After a day with many session restarts - When investigating tool-call ceiling hits
Related Skills
llm-wiki-audit-feedback-loop
Durable feedback loop for correcting llm-wiki pages without losing the correction to chat history. Use when (1) a human notices a wiki page is wrong, outdated, or contradicts a source, (2) processing the `audit/` inbox of a domain wiki, (3) reviewing what feedback has been resolved vs deferred, (4) needing to leave a comment on a specific text range that survives line- number drift. Implements the anchored-text audit file pattern from lewislulu/llm-wiki-skill, adapted for workspace-hub's domain-wiki layout under /mnt/local-analysis/llm-wiki/wikis/<domain>/. Extends the 5-op model (compile/ingest/query/lint) from research/llm-wiki with the missing `audit` op. Never silently delete feedback — rejected audits stay archived with rejection rationale.
pre-completion-cleanup-audit
Audit and dispose of session residue (orphan files, scratch dirs, sibling-repo state, locks, trash-stages) BEFORE claiming a task complete. Required gate before any agent says "all done", "task complete", or hands work back to user/orchestrator.
repo-mission-portfolio-audit
Audit the workspace-hub repo portfolio to extract each repo's mission, identify documentation gaps, and prioritize a plan/approval sequence with explicit LLM-wiki weighting for future issue triage.
provider-session-ecosystem-audit-and-exporters
Build and maintain cross-provider session-log audits for Codex, Codex, Hermes, and Gemini, including exporter design, normalization, and behavioral verification.
orcawave-orcaflex-readiness-audit
Audit the real readiness of digitalmodel OrcaWave/OrcaFlex spec-driven workflows by reconciling workspace-hub issues, source/tests, semantic-equivalence boundaries, and wiki synthesis gaps.
tax-filing-session-setup-with-github-tracking
Structured workflow for preparing and tracking a tax filing session using prepared documents, task checklist, and GitHub issue cross-referencing
tax-filing-session-setup-with-github-traceability
Structured workflow for setting up a multi-file tax filing session with GitHub issue tracking and prepared-file validation
read-only-pre-implementation-audit
Systematic cross-check workflow to validate assumptions before TDD coding begins
handle-session-overlay-blocking
Technique for dismissing overlay dialogs that freeze rendering and block form interaction in web automation
handle-freetaxusa-session-timeouts
Recover from FreeTaxUSA session timeout dialogs blocking form submission and navigation
extract-skills-from-hermes-sessions
Automatically analyze Codex session transcripts to identify and extract reusable skills using LLM analysis via OpenRouter
extract-skills-from-Codex-sessions
Automatically extract reusable skills from Codex session transcripts using LLM analysis and wire them into a Stop hook