session-metrics

Tally Claude Code session token usage and cost estimates from the raw JSONL conversation log. Trigger when the user asks about session cost, token usage, API spend, cache hit rate, input/output tokens, or wants a breakdown of how much a Claude Code session has cost. Also trigger for "how much have we spent", "show me token usage", "session summary", "cost so far", or any request to analyse or display per-turn metrics from the current or a past session. Do NOT auto-dispatch compare mode (--compare / --compare-prep / --compare-run / --count-tokens-only) from natural-language phrases. The skill body uses $ARGUMENTS[0] as the dispatch key — if the first positional argument is not literally "compare", "compare-prep", "compare-run", or "count-tokens", route to the default single-session report.

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

75%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A highly actionable, well-structured reference with excellent dispatch routing and real bundle-backed progressive disclosure. Its weakness is token weight: version-numbered feature history, UI interaction detail, and an oversized example table inflate the body beyond what an overview needs.

Suggestions

Move version-numbered feature notes (v1.6.0 cross-cutting sections, v1.7.0 Phase-B attribution, v1.36.0 share-safe, v1.78.0 insights, v1.87.0 model column) into a short changelog section or the relevant reference file so the body reads as current behavior, not history.

Defer the HTML-only UI detail (drawer interaction, truncation caps, collapsible sections) to references/jsonl-schema.md or an HTML-format reference, keeping only the flag that controls it inline.

Trim the 12-row export example table to 3–4 representative cases (session→html, project→html, project→html csv, all-projects) — the ordered scope rules already make the rest derivable.

Dimension	Reasoning	Score
Conciseness	The body is dense and mostly functional, but noticeably padded in places: HTML UI interaction detail (drawer Esc/backdrop close, focus return, ~240-char truncation caps), a 12-row export-example table, and version-numbered feature notes (v1.6.0, v1.7.0, v1.36.0, v1.78.0, v1.87.0) sprinkled inline rather than confined to a changelog/deprecated section. Fits the 'mostly efficient but could be tightened' anchor better than the 'only minor trimming needed' one.	3 / 5
Actionability	Fully executable guidance throughout: copy-paste `uv run python …` commands, a literal-equality dispatch table on `$ARGUMENTS[0]`, ordered first-match-wins export rules, an arg-string→exact-command example table, and a complete flag reference. Matches the copy-paste-ready anchor.	5 / 5
Workflow Clarity	Multi-step flows are clearly sequenced with explicit gates (tasks companion: prepare → edit → render; insights: prepare → digest → render; scope gates for the companion; `--prune-exports` dry-run-by-default with `--yes` for the destructive path). Falls short of anchor 5 because there are no explicit error-recovery/feedback loops if a run or render step fails.	4 / 5
Progressive Disclosure	Clear section structure, well-signaled one-level-deep references that all exist in the bundle (model-compare.md, instance-dashboard.md, jsonl-schema.md, pricing.md, custom-prompts.md, platform-notes.md, post-export-audit.md, tasks-companion.md), and deferral that is explicitly motivated. Held at anchor 4 because the ~470-line body inlines substantial detail (subagent-attribution internals, HTML drawer/section behavior, full column semantics) that could live in references/jsonl-schema.md.	4 / 5
	Total	16 / 20 Passed

Description

95%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description: concrete capability statement, rich natural trigger vocabulary, explicit when-guidance, and an unusual but valuable negative boundary clause. The only soft spot is that it is somewhat long for a description and leaves secondary capabilities (exports, dashboards) implicit.

Dimension	Reasoning	Score
Specificity	Names the domain and several concrete actions — "Tally Claude Code session token usage and cost estimates from the raw JSONL conversation log", per-turn breakdowns, "cache hit rate" — but capabilities like export formats and multi-session/project dashboards go unmentioned, so coverage has minor gaps rather than being comprehensive.	4 / 5
Completeness	Explicitly answers both questions: what ("Tally … token usage and cost estimates from the raw JSONL conversation log") and when ("Trigger when the user asks about …", "Also trigger for …") with concrete trigger phrases, mirroring the anchor-5 example.	5 / 5
Trigger Term Quality	Comprehensive natural-language coverage: "session cost", "token usage", "API spend", "cache hit rate", "input/output tokens", plus quoted phrases users would actually say ("how much have we spent", "show me token usage", "cost so far"). Matches the comprehensive-synonyms anchor.	5 / 5
Distinctiveness Conflict Risk	Clear niche (Claude Code session cost/token metrics) with distinct triggers, and an explicit negative boundary ("Do NOT auto-dispatch compare mode … from natural-language phrases") that further reduces mis-routing risk.	5 / 5
	Total	19 / 20 Passed

Validation

81%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 1 deeper-than-1-level	Warning
referenced_paths_exist	Referenced path issues: 2 deeper-than-1-level	Warning

	Total	13 / 16 Passed

Repository: centminmod/my-claude-code-setup
Path: .claude/skills/session-metrics/SKILL.md
Commit: f93acbc

Reviewed: 8 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.