gtapps/cost-reflect
plugins/claude-code-hermit/skills/cost-reflect/SKILL.md
Use when the user asks about cost structure or composition: which token types cost the most, how cache reads compare to output costs, which sessions are draining budget, or what cold starts add up to. This is the 'why is my bill high' skill. It audits where spend goes across token categories (cache_read, cache_write, output, input) and per-session attribution. Invoke for questions like 'cost breakdown', 'which sessions are expensive', 'what's driving spend', or 'cold start costs'. Not for week-over-week trend lines (use hermit-evolution) or simple total-spend summaries.
$ install --global
skillsauthnpx skillsauth add gtapps/claude-code-hermit cost-reflectInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: Jun 12, 2026, 4:35 AM192.5s1 file scanned
SKILL.md
- name:
- cost-reflect
- description:
- Use when the user asks about cost structure or composition: which token types cost the most, how cache reads compare to output costs, which sessions are draining budget, or what cold starts add up to. This is the 'why is my bill high' skill. It audits where spend goes across token categories (cache_read, cache_write, output, input) and per-session attribution. Invoke for questions like 'cost breakdown', 'which sessions are expensive', 'what's driving spend', or 'cold start costs'. Not for week-over-week trend lines (use hermit-evolution) or simple total-spend summaries.
Cost Reflect
Runs a structural cost audit over the last 7 days of cost-log.jsonl and delivers the result to the operator. The math is done by a script — your job is to run it and deliver the output through the right channel.
Step 0 — Channel reply
If this skill was invoked from a channel-arrived message (the inbound prompt contains a <channel source="..."> tag), reply via that channel's reply tool. Otherwise emit to conversation.
Step 1 — Run the analyzer
bun ${CLAUDE_PLUGIN_ROOT}/scripts/cost-reflect.ts .claude-code-hermit
Capture stdout. If the output starts with "No cost data", report that message and stop — there's nothing to send.
Step 2 — Relay the report
The script output is already formatted and capped at 1500 characters. Relay it verbatim. Don't summarize, reformat, or add commentary — the output is the report.
Step 3 — Channel delivery
When this skill fires from a routine or proactively (i.e., not from a direct operator conversation), deliver the report through the operator's channel:
-
Resolve the outbound channel:
bun ${CLAUDE_PLUGIN_ROOT}/scripts/resolve-outbound-channel.ts .claude-code-hermitParse stdout as JSON. On success (
"id"and"chat_id"present), send viamcp__plugin_<id>_<id>__replywith{ chat_id, text: <report> }where<id>is the resolved channel name. -
If the script exits non-zero or returns
{"error":"no_reachable_channel"}: ifpush_notifications === trueinconfig.json, firePushNotification(message="cost-reflect: <first line of report>", status="proactive")so the summary still reaches the operator. Then append a single Findings line to.claude-code-hermit/sessions/SHELL.md:"cost-reflect: no reachable channel configured, channel-send skipped". Only log this once per session to avoid noise. -
To set a preferred channel, add
"primary": "<channel-name>"insidechannelsinconfig.json.
Notes
- Scheduling: this skill doesn't self-register a routine. To run it weekly, add it via
/claude-code-hermit:hermit-settings— a Sunday 22:00 cadence (0 22 * * 0) before weekly-review works well. - What it measures: token-type cost composition (cache_read / cache_write / output / input), cold-start turns (context warm-ups with no prior cache hit), and per-session cost attribution. For week-over-week totals and autonomy trends, use
/claude-code-hermit:hermit-evolutioninstead.
Related Skills
gtapps/session-start
data-ai
VerifiedTrustedCommunity
Initializes or resumes a work session. Loads context from OPERATOR.md and SHELL.md, orients the agent, and establishes what to work on. Use at the beginning of every work session.
62SKILL.mdUpdated Apr 27, 2026
gtapps/hermit-evolve
tools
VerifiedTrustedCommunity
Evolves hermit configuration and templates after a plugin update. Detects version gaps, presents new features, walks through new settings. Run after updating the plugin.
62SKILL.mdUpdated Apr 27, 2026
gtapps/hatch
testing
VerifiedTrustedCommunity
Initializes the autonomous agent in the current project. Creates the state directory, templates, OPERATOR.md, and config.json. Appends session discipline to CLAUDE.md. Detects installed hermits. Run once per project, like git init.
62SKILL.mdUpdated Apr 27, 2026
gtapps/docker-setup
tools
VerifiedTrustedCommunity
Generates Docker scaffolding and walks the operator through the full deployment — token setup, build, start, MCP plugin configuration, workspace trust, and verification. Offers to back up and overwrite existing Docker files. Run after /hatch.
62SKILL.mdUpdated Apr 27, 2026