Session Scan Skill for Claude
Compute metrics for Claude Code sessions. Discovers via ccrider, filters trivial, computes friction/opportunity/fingerprint scores. Use for broad session triage.
Install
one command, takes just this skill from the repositorynpx skills add https://github.com/oliver-kriska/claude-elixir-phoenix --skill session-scanWhat comes with it
72 684 bytes besides the instructionWhat it tells the agent to use
found in the instruction textThe instruction itself
15 sections, as written by the authorSession Scan (Tier 1)
Compute deterministic metrics for sessions discovered via ccrider MCP.
STOP — Read Before Executing
NEVER call mcp__ccrider__get_session_messages in main context.
Each response is 5-50KB of JSON. Fetching N sessions directly = context crash.
You MUST delegate all scoring to subagents. The main context only does:
discovery, deduplication, subagent spawning, and result collection.
Requirements
Requires ccrider MCP. If not available:
> ccrider MCP is required. See: <https://github.com/neilberkman/ccrider>
Usage
/session-scan # Last 7 days, up to 50 sessions
/session-scan --since 2026-02-01 # Since specific date
/session-scan --project enaia # Filter by project name
/session-scan --limit 20 # Cap session count
/session-scan --list # Discovery only, no scoring
/session-scan --rescan # Recompute already-scanned sessions
What Main Context Does
1. Parse Arguments
Extract from $ARGUMENTS:
--since DATE: ISO date filter (default: 7 days ago)--project NAME: Filter by project path substring--limit N: Max sessions to process (default: 50)--list: Discovery only — show table, skip scoring--rescan: Recompute metrics for already-scanned sessions
2. Discover Sessions (safe — response is ~1KB)
mcp__ccrider__list_recent_sessions(limit: N, project: PROJECT, after_date: SINCE)
Filter out sessions with < 10 messages. If --list: show table and stop.
3. Deduplicate
Read .claude/session-metrics/metrics.jsonl (if exists).
Skip sessions already scanned unless --rescan flag is set.
Report: "N new sessions to scan (M already in ledger)"
4. Resolve Scorer Path
Glob: **/session-scan/references/compute-metrics.py
Store as SCORER_PATH (absolute). PROJECT_ROOT = current working directory.
5. Spawn ONE Subagent Per Session
DO NOT fetch messages yourself. DO NOT run Python yourself.
You MUST use the Task tool to spawn subagents.
For EACH unscanned session, spawn a haiku subagent using this template:
Task(subagent_type="general-purpose", model="haiku", mode="bypassPermissions", prompt="""
Score one Claude Code session. Steps:
1. Fetch messages:
mcp__ccrider__get_session_messages(session_id: "{SESSION_ID}", last_n: 200)
2. Write the full tool result (messages JSON) to:
{PROJECT_ROOT}/.claude/session-metrics/_tmp_{SHORT_ID}.json
3. Run scorer and capture result in ONE Bash command:
python3 {SCORER_PATH} {PROJECT_ROOT}/.claude/session-metrics/_tmp_{SHORT_ID}.json --session-id {SESSION_ID} --project {PROJECT_NAME} > {PROJECT_ROOT}/.claude/session-metrics/_result_{SHORT_ID}.json && rm {PROJECT_ROOT}/.claude/session-metrics/_tmp_{SHORT_ID}.json
4. Report back: "Scored {SHORT_ID}: friction=X.XX fingerprint=Y"
If ccrider fails, report the error and exit.
""")
Spawn ALL subagents in parallel. Wait for all to complete.
6. Collect Results
for f in .claude/session-metrics/_result_*.json; do
cat "$f" >> .claude/session-metrics/metrics.jsonl
rm "$f"
done
7. Display Results
Show summary table sorted by friction (descending):
| ID | Project | Date | Type | Friction | Opportunity | Tier2? |
|----------|---------|------------|-------------|----------|-------------|--------|
| ffa155ee | enaia | 2026-02-18 | bug-fix | 0.42 | 0.65 | Yes |
| 90a74843 | wedding | 2026-02-17 | feature | 0.15 | 0.20 | No |
If Tier 2 eligible: suggest /session-deep-dive --from-scan.
8. Update Scan Metadata
Write .claude/session-metrics/latest-scan.json:
{
"scanned_at": "2026-02-20T14:30:00Z",
"sessions_discovered": 25,
"sessions_scanned": 18,
"sessions_skipped": 7,
"tier2_eligible": 5
}
Scoring Reference
See references/scoring-guide.md for full algorithm documentation.
- Friction Score (0-1): Weighted signals, sigmoid-normalized
- Fingerprint: Rule-based classifier (6 types)
- Plugin Opportunity (0-1): Missed
/phx:commands - Tier 2 Eligible: friction > 0.35 OR opportunity > 0.5 OR plugin used OR msgs > 50
Iron Laws
- NEVER call
get_session_messagesin main context — WILL crash context - ONE subagent per session — each fetches exactly ONE session
mode="bypassPermissions"on every Task — subagents need Write+Bash- Absolute paths in subagent prompts — subagents don't inherit context
- Result files, not direct append — avoid jsonl race conditions
- NEVER modify existing metrics.jsonl entries — append-only ledger
- ALWAYS delete temp and result files after processing
Other skills for the same job
different authors, same section of the catalogue
Meta-skill for publication-ready figures. Use when creating journal submission figures requiring multi-panel layouts, significance annotations, error bars, colorblind-safe palettes, and specific journal formatting (Nature, Science, Cell). Orchestrates matplotlib/seaborn/plotly with publication styles. For quick exploration use seaborn or plotly directly.
Coding Agent Account Manager - Sub-100ms account switching for AI coding CLIs with fixed-cost subscriptions. Vault profiles, isolated profiles for parallel sessions, smart rotation with health scoring, cooldown tracking, automatic failover, TUI dashboard. Go CLI.
Use this umbrella skill when the request spans multiple Cloud Claw user-facing domains, especially launching a new AltClaw or OpenClaw VM and then managing lifecycle, logs, renewal, or dashboard access through the local altllm cloud-claw-* commands in this repository.
Creating interactive data visualisations using d3.js. This skill should be used when creating custom charts, graphs, network diagrams, geographic visualisations, or any complex SVG-based data visua...
Use this skill when you need to report on a troubleshooting session
Query and browse evaluation results stored in MLflow. Use when the user wants to look up runs by invocation ID, compare metrics across models, fetch artifacts (configs, logs, results), or set up the MLflow MCP server. ALWAYS triggers on mentions of MLflow, experiment results, run comparison, invocation IDs in the context of results, or MLflow MCP setup.
Guided journey from an app people sign up for and then quietly abandon to a sealed retention engine with a habit loop, an activated first run, and one metric the whole team trusts. Orchestrates eight skills phase by phase - hooked-ux, improve-retention, continuous-discovery, lean-ux, inspired-product, lean-analytics, microinteractions, drive-motivation - asking the user questions at every decision point and recording results in the project docs/ folder (PRODUCT.md, METRICS.md, GROW-APP-PLAN.md) so the journey resumes across sessions. Use when the user wants to lift activation and retention, design a habit loop, fix a leaky onboarding funnel, or says ''users sign up then disappear''. Do not use to fix broken UX or performance that no engagement mechanic can paper over - run improve-app first; if there is no app yet, use create-app. For one framework in isolation, invoke that skill directly.
Guided journey from a stalled, plateaued business to one with an honest diagnosis, a working operating rhythm, and offers repriced to real value. Orchestrates eight skills phase by phase - good-strategy-bad-strategy, traction-eos, high-output-management, team-topologies, drive-motivation, lean-analytics, negotiation, monetizing-innovation - asking the user questions at every decision point and recording results in the project docs/ folder (STRATEGY.md, OPERATIONS.md, METRICS.md, IMPROVE-BUSINESS-PLAN.md) so the journey resumes across sessions. Use when the user wants to fix a business that has plateaued, diagnose why growth stalled, tighten strategy and execution, re-motivate a team, or says ''revenue is flat and I do not know why''. Starting from scratch with no customers: use create-business. Once the fundamentals work and the goal is expansion: use grow-business. When the product itself drags the business down: use improve-app. For one framework in isolation, invoke that skill directly.
How to use it
Take oliver-kriska/session-scan from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.